Research ArticleWearable Sensors for Measuring Movement in ShortSessions of Mindfulness Sitting Meditation A Pilot Study

Victor H Rodriguez 1 Carlos T Medrano 12 and Inmaculada Plaza 12

1EduQTech EUPT Universidad de Zaragoza cAtarazana 2 44003 Teruel Spain2IIS Aragon Universidad de Zaragoza Zaragoza Spain

Correspondence should be addressed to Carlos T Medrano ctmedraunizares

Received 28 December 2017 Revised 27 March 2018 Accepted 15 April 2018 Published 7 May 2018

Academic Editor Jesus Favela

Copyright copy 2018 Victor H Rodriguez et al +is is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work isproperly cited

Mindfulness techniques are useful tools in health and well-being To improve and facilitate formal training beginners need toknow if they are in a stable sitting posture and if they can hold it Previous monitoring studies did not consider stability duringsitting meditation or were specific for longer traditional practices In this paper we have extended and adapted previous studies tomodern mindfulness practices and posed two questions (a) Which is the best meditation seat for short sessions In this way theapplications of stability measures are expanded to meditation activities in which the sitting posture favors stability and (b)Whichis the most sensitive location of an accelerometer to measure body motion during short meditation sessions A pilot studyinvolving 31 volunteers was conducted using inertial sensors+e results suggest that thumb head or infraclavicular locations canbe chosen to measure stability despite the habitual lumbar or sacral region found in the literature Another important finding ofthis study is that zafus chairs and meditation benches are suitable for short meditation sessions in a sitting posture although thezafu seems to allow for fewer postural changes +is finding opens new opportunities to design very simple and comfortablemeasuring systems

1 Introduction

Around 1980 Jon Kabat-Zinn adapted some meditationskills originally from oriental traditions provided them witha lay and scientific character and applied them to reducingstress [1] Kabat-Zinn defined mindfulness as the awarenessthat emerges through paying attention on purpose in thepresent moment and nonjudgmentally to the unfolding ofexperience moment by moment [2] Since then mindfulnesshas been increasingly applied in health and well-being [3ndash6]

Although there are two broad schools of mindfulness (atraditional Buddhist approach and a contemporary westernapproach) embodiment is a common process that can in-tegrate them [7] For mindfulness practice the experience ofbeing embodied involves paying attention to the body [8]+e observing part of body mindfulness is related to bodyawareness defined as ldquothe subjective phenomenologicalaspect of proprioception and interception that entersconscious awarenessrdquo [9] Body awareness is a key process

in understanding the differences between meditators andnonmeditators [10]

Mindfulness manuals tend to recommend beginningwith the daily practice of mindfulness of breath duringformal seated meditation sessions [11] Although sometechniques recommend daily practice of 45ndash60minutes newstudies recommend shorter sessions between 5 and 15minutes in the beginning stages [12] Buddhist traditiontexts describe the proper sitting posture the back should bekept straight and the legs crossed Similarly sitting medi-tation is a basic technique in western mindfulness medi-tation [13] A relevant aspect of the meditation posture isstability and this can be achieved sitting in an uprightposture that embodies dignity [12 14] Although a relaxedposture is adopted prolonged motionlessness can lead topain in muscles and joints Mindfulness meditation in-structors often encourage students not to shift position torelieve the pain but instead to focus careful attention directlyon the pain sensations and to assume a nonjudgmental

HindawiJournal of Healthcare EngineeringVolume 2018 Article ID 7275049 9 pageshttpsdoiorg10115520187275049

attitude toward these sensations to reduce the distress [3]Recently Jones [15] investigated the relationship betweenthe posture and outcomes of mindfulness practice Partic-ipants rated how comfortable they found each posture alongwith the difficulty to maintain it +is pilot study concludedtentatively that distress tolerance seemed to decrease in theslouched posture Qualitatively it appeared that the uprightposture may facilitate breathing during mindfulness prac-tice Researchers could widen their focus to examine the roleof the body in bringing about the beneficial outcomes ofmindfulness-based interventions [15]

+ere are several kinds of meditation seats Orientalmeditation is performed while seated on a cushion in eitherthe full-lotus or half-lotus position [16] Instead of a cushion(zafu) a comfortable chair can be used for people with backpain or difficulty getting up [17] Researchers have also useda kneeling meditation bench that is especially helpful forbeginners to optimize spinal alignment and reduce weightand stress on the knees hips ankles and back [18]

Monitoring the mindfulness practice is not a novel ideaand several studies measure physiological variables duringmeditation Ahani et al [19] recorded electroencephalo-graphic (EEG) and respiration signals during a mindfulnessmeditation intervention Arch and Craske [20] followeda focused breathing induction in which they controlled theheart rate (beats per minute) In a similar way Vidyarthi andRiecke used two breathing sensors which measurethoracicabdominal expansion Custom algorithms werecreated to extract parameters from the data respiratorydepth respiratory length and thoracic-to-abdominal ratio[21] However few of the previous studies considered sta-bility or posture In this regard accelerometers have longbeen used to measure the posture [22 23] or stability[24ndash27] In the clinical context stability has been referred toas a personrsquos ability to maintain the position of the body ormore specifically its center of mass within specificboundaries [28] For example stability of constructionworkers [26] older people [27] or people with Parkinsonrsquosdisease [25] has been measured generally under conditionsthat challenge stability (stance on foam with eyes closed orposture with loads) In the present study stability is un-derstood in a more general perspective and it is consideredto be equivalent to lack of motion Ideally the meditationposture favors stability Chang et al were pioneers [29 30] inthis topic and proposed measuring vibration degrees asa good index in training body stillness A lower motionwouldmean a better meditation state+ese authors proposea body stillness monitoring system based on a triaxial ac-celerometer Both a mean motion index and a maximummotion index were derived from the square summation ofthree axes In a posterior study [31] they used the meanmotion index that was extracted from accelerometers placedon the arm and on the chest in controlled wide amplitudemovement experiments It was determined that the arm wasa more sensitive location +en the sensor on the arm wasused during Chan Ding practices +e mean motion indexwas shown to be different for inexperienced people

Chang et al [30] secured the measurement system usinga belt between the abdomen and chest or on the arm [31]

Nevertheless other authors describe other body locations(head sternum lumbar or sacral region wrist etc) tomeasure postural sway stability or balance especially whilestanding [32 33] A review of commercial products in therelated problem of postural control indicates the same va-riety of body positions for the wearable sensor collarbonethe lower back the back of the neck or glasses [34ndash37]

+e current pilot study is motivated by the possibility ofmeasuring motion and postural changes in mindfulnesssitting meditation +is would allow users to receive feedbackon their progress and to facilitate further investigations aboutthe role of movement and posture in the outcomes ofmindfulness-based interventions We will focus on two fac-tors the meditation seat and the sensor location +e con-tributions of this paper come from the answer to the followingtwo research questions (i) Which is the seat that most favorsstability In this way we are expanding the range of appli-cations of stability measures to meditation activities in whichthe sitting posture should favor stability and low range ofmovements and (ii) Which is the most sensitive location forinertial sensors to measure body motion Our study differsfrom the work of Chang et al [30 31] in the characteristics ofthe meditation (beginner-oriented only 10min no full-lotusor half-lotus required) in the inclusion of more sensor lo-cations and the analysis of the influence of the seat

2 Methods

21 Participants +e sample consisted of 31 participants 16males and 15 females with a mean age of 288 (range 18ndash46)Out of the participants 27 had no experience and four hadprevious experience in mindfulness meditation rangingfrom 4 to 20 years

+e study protocol was approved by the Committee ofEthics for the Clinical Investigation of Aragon (CEICA) Allthe subjects received the information about the study in oraland written format and they approved it

22Procedure According to the literature review performedat the beginning of this work several seats for sitting duringmindfulness meditation were tested a chair a zafu (a smallcushion) and a meditation bench (Figure 1)

+e subjects were required to perform three times a shortbreathing meditation session one for each type of seat +eorder of the seats was randomly assigned but compensatedso that the three seats were approximately selected equallyfor the first second and third sessions Nobody chose a lotusposture (full or half) with the zafu Each session lasted 10minutes +e participants relaxed and rested 2 minutesbetween sessions (or longer in case of pain) Most mind-fulness practices began with the observation of breathing[38] so the volunteers were asked to focus on their breathand to count groups of five breathing cycles (inhalation-exhalation) Whenever their attention wandered from theirbreath to inevitable thoughts and feelings that could arisethe subjects simply had to take notice of them and then letthem go as attention was returned to breathing using theirbreath as an anchor [13]

2 Journal of Healthcare Engineering

During the meditation sessions a series of inertial sensorswere placed on dierent body locations as described in Figure 2

More specically the sensors were placed as follows

(1) An IMU9150 [39] (plusmn4G 16-bit resolution) on thehand specically on the left thumb fastened withstrips (from now on this location will be referredto in short as thumb)

(2) An IMU9150 on the head held by glasses (inshort head)

(3) An ADXL345 [40] (plusmn4G 13-bit resolution) on thexiphoid process of the sternum (in short sternum)

(4) An IMU9150 on the left infraclavicular region onthe pectoralis major just below the collarbone (inshort infraclavicular region)

(5) A smartphone (Samsung Galaxy Trend Plus GT-S7580 accelerometer with plusmn2G 12-bit resolu-tion) on the left side of the lumbar inside a specialbag (in short lumbar region)

(a) (b) (c)

Figure 1 Meditation seats used in the short meditation sessions (a) chair (b) zafu (c) meditation bench

Head (held by glasses)(IMU9150)

Infraclavicular region(IMU9150)

Sternum(ADXL345)

Le side of the lumbar(smartphone (GT-S7580))

Le thumb(IMU9150)

Figure 2 Schematic location of sensors on the body

Journal of Healthcare Engineering 3

e sensors were selected for convenience availability inour laboratory and ease of use with libraries for the Arduinoplatform available although many commercial parts couldhave been selected too ey fulll the requirements of theapplication (Section 221) Since a smartphone was used togather data (as it would be in a real application of the system)we took advantage of its accelerometer to add another lo-cation (lumbar region) ere was no relation between thesensor and location other than the convenience of placing thesmartphone in the lumbar region with a commercial bag

221 Signal Acquisition Signals were acquired through anArduino Pro Mini [41] which gathered information from foursensors (inertial sensor) with a frequency of 50Hze librariesof the inertial sensors were provided by SparkFun makinguse of the I2C protocol for communication e IMU9150only has two addresses in the same I2C bus Since our kit hadthree IMU9150 a second I2C bus was required e ArduinoPro Mini has only one built-in I2C bus (hardware imple-mentation)erefore the second onewas emulated by softwareAll sensor information was sent in real time to a smartphonethrough an HC-05 Bluetooth interface [42] at a rate of 115200bauds is serial communication was the bottleneck in thetransmission Sensor data were transmitted as ASCII char-acters and the maximum transmission frequency with thatserial port speed was 68Hz given the number of charactersper sampling period in our experiment higher than the selectedfrequency While the smartphone collected informationfrom the Arduino it sensed meditator movement with itsinternal accelerometer at a frequency of 50 Hz

e frequency was selected according to the followingreasoning Basically the movement is expected to be underalmost static conditions thus a low-frequency measurementEven normal movements like gait are of low frequency (gait ismeasured in 06ndash5Hz in [43]) One could also consider thepossibility of tremor induced by the posture Enhancedphysiological tremor and even pathological tremor are below18Hz [44] and other studies have reported detecting tremorcaused by fatigue in a low frequency of 2ndash6Hz in [45] or 2Hzpeak in [46]us a sampling frequency of 50Hz was enoughand allowed a comfortable wireless Bluetooth transmissionMinimal data losses were reported in our experiments andthe corresponding values were found by interpolation in thepostprocessing Data losses could be due to transmissionerrors or due to errors when storing data in les Finallyexperimentally it seems that 50Hz corresponds to one of thesampling options in the selected smartphone model (optionldquoSENSOR_DELAY__GAMErdquo in the Android source code[47]) with some variability since this device cannot deal withstrict real time and the values have to be interpolated too

After every meditation session the information stored ina le was labeled with the name of the seat (chair zafu andmeditation bench) and the subjectrsquos identier

222 Signal Processing e les were processed withMATLAB e rst 60 and last 30 seconds of each sessionwere removed to avoid incorrect data coming from peoplesitting down or from the smartphone being removed from

the special bag Two types of processing were performedFirst the acceleration norm for each sensor was calculatedand then its standard deviation (σa) during each meditationsession was extracted is value is considered a conven-tional measure of stability that is sensitive to small and rapidchanges in the posture such as tremors due to not beingable to maintain the posture

Since parts of dierent manufacturers were used theprocessing was repeated by discarding the least signicantbits of the most accurate sensors In this way we simulateda situation in which all the sensors had the same resolution(the worst one) and we tested the inordfuence on the results

Subsequently the accelerometer signal was processed toobtain a global measure of changes in the volunteerrsquos positione idea was to distinguish major changes during the sessionsey could be caused by volunteers adopting a dierentposition due to their inability to hold the previous one or dueto slow posture variations For this purpose the study focusedon the sensor placed on the head Specically the sensor wasplaced on the temple of the glasses since this location allowsfor an easy interpretation of deviations from the uprightposition in terms of acceleration components e approxi-mate orientation of the sensor is as shown in Figure 3 It isreasonable to assume that the acceleration is mainly due togravity because the movements while meditating are slowBesides orientation values obtained from accelerometerreadings were subjected to a low-pass ltering process (see (2))to enhance their reliability As shown in Figure 3 the maincomponent of the acceleration is along the x-axis Tilts aroundthe left-right direction or around the anterior-posterior di-rection led to an increase of y and z components respectively

us the angles of the acceleration with respect to theplanes xz and xy were measured using the followingequations

Axz tanminus1aya2x + a2zradic

Axy tanminus1az

a2x + a2yradic

(1)

Figure 4 shows the two measured angles e Axz anglecorresponds to an anterior-posterior movement and theAxy angle corresponds to a left-right movement

Gravity force

x

y

Figure 3 Approximate orientation of the sensor with respect to theEarthrsquos gravity force

4 Journal of Healthcare Engineering

After recording the angles they were filtered usinga moving average filter (500 samples 10 s) to reduce noise((2) and Figure 5) which effectively removed any dynamiccomponent

y(n) 1N

1113944

iNminus1

i0x(nminus i) (2)

where N 500A global measure of the volunteerrsquos change of position

during the session was then obtained from the differencebetween the maximum and minimum values of the signalFor example the signal in Figure 5 has a variation of 1245deg

223 Statistical Analysis +e standard deviation of theacceleration was analyzed using a two-way ANOVA withrepeated measures on both factors the sensor location andthe seat (results in Section 31) Similarly to analyze theglobal change of the posture (angle differences) an ANOVAwith one factor (seat) was carried out since in this secondcase only one sensor location was used (Section 32) +estatistical analyses were performed with SPSS 120

3 Results

31 Analysis of Accelerometer Standard Deviation Twofactors have been considered in the analysis the meditationseat during the sessions (ldquoseatrdquo chair zafu or meditationbench) and the location of sensors on the body (ldquosensorlocationrdquo thumb head sternum infraclavicular region orlumbar region)

Table 1 shows the results of an ANOVA with repeatedmeasures on both factors +e details about the statisticaltests performed are given in Appendix +e sensor locationwas found to be significant (p valuelt 005) On the otherhand neither the seat nor the interaction term was signif-icant Since the sensor location was found to be significantwe proceeded with a multiple comparison analysis to dis-cover the root of the differences in terms of pairwisecomparisons of sensor locations +e p values of thecomparisons are shown in Table 2 after applying a standardBonferronirsquos correction

Table 3 shows the mean values obtained for each of thelocations of the sensors and Table 4 shows the mean valuesobtained for each of the seats (marginal values provided by

SPSS) +e marginal values are obtained by averaging overone of the factors keeping only the other one +ey can givea clue of the influence of a single factor According to Ta-bles 2 and 3 the lumbar position is the least sensitive since itsdifferences with the others are significant and themean valueis the lowest one Among the remaining positions (thumbhead sternum and infraclavicular region) the thumb showsthe highest sensitivity although the interdifferences are notsignificant except with the sternum+e marginal values forthe seat factor in Table 4 are presented for completeness andto show that the experimental values are very close

+e results of this statistical analysis remained almostthe same after preprocessing the accelerometer to lower theresolution to the worst one as explained in Section 222+e only difference is that sternum and thumb did not differsignificantly after that preprocessing

32 Posture Variation Depending on the Seat +e results ofthe ANOVA for the anterior-posterior sway (Axz) and theleft-right sway (Axy) are presented in Table 5 Table 6 showsthe values of the sway for the three seats

+e zafu seems to permit a better posture setting that canbe held throughout the session since sways were lowerHowever for Axz the effect was not significant while forAxy it was +erefore a post hoc analysis was performed forAxy +e results are presented in Table 7 +e pairwisecomparisons indicate that the zafu seat was different fromthe others

4 Discussion

Postural stability is an important factor during mindfulnessmeditation training Inertial sensors can be a good option formonitoring posture dynamics due to their low cost and easeof use One of the goals of this paper was to determine whichis the most sensitive sensor location Multiple protocols weredescribed in the literature for the placement of wearablesensors on the human body for assessing for instancestanding balance or walking stability Most of them reportedplacing a wearable sensor on either the lumbar or sacralregion of the trunk To evaluate body posture stability duringsitting meditation Chang et al [30] secured the measure-ment system between the abdomen and the chest or on thearm [31] Manufacturers of commercial products choseother body positions collarbone head lower back and soon +e lumbar region is often selected since it is close to thebody center of mass in a standing posture However theresults of the present study indicate that the lumbar or sacralregion is not the best option for measuring motion duringsitting meditation since its σa value is the lowest and sig-nificantly different from the rest It could be ruled out forfuture applications Among the remaining locations it is notpossible to separate them into different groups and only thethumb is significantly different from the sternum+ereforeconvenience or comfort can be important factors to selectthe location +e head location requires a more bulky ac-cessory +e infraclavicular region is less invasive since thesensor is attached to clothes +e sternum and lumbar could

Angle Axy

Angle Axy

Angle Axz

Angle Axz

Figure 4 Direction of rotation angles with respect to glasses Bluearrow anterior-posterior (Axz) white arrow left-right (Axy)

Journal of Healthcare Engineering 5

be more accepted by users since sensors attached to strapsare common in commercial products for tness Finally thethumb location would require a miniaturization of thesystem to improve comfortability Usability acceptance andcomfort should be evaluated among users Although thedierences are not statistically signicant the sensitivity of

the head location was higher than the sensitivity of thesternum or the infraclavicular locations is seems rea-sonable since the distance to the seat base is larger on thehead In fact the order of the sensitivity among the locationshead infraclavicular region sternum and lumbar regionfollowed the order of the distance to the seat e thumbshowed an even larger degree of movement but this could berelated to the movements of the hand that can be in-dependent of upper body displacements In fact handposture is very important according to traditional texts[48 49] and in particular thumb ngers should not drop orrise but remain in a perfect horizontal position [50]

e other goal of this paper was to analyze the inordfuenceof the seat to recommend a specic one based on dataobtained using inertial sensors In the oriental traditionmeditation is performed while seated on a cushion pref-erentially in the full-lotus or half-lotus position e staticposition is not to be changed until the end of the meditation

Table 1 ANOVA results for σa (p values)

Factor p valueSeat 0744Sensor location 0000Seatlowast sensor location 0726p values less than 005 are in bold

Table 2 Sensor location analysis pairwise comparisons withBonferronirsquos correction

umb Head Sternum Infraclavicularregion

Head 1000 mdash mdash mdashSternum 0022 0329 mdash mdashInfraclavicularregion 0787 1000 1000 mdash

Lumbar region 0000 0000 0000 0000p values less than 005 are highlighted in bold

Table 3 Sensor location marginal mean values of σa (in g)

Sensor location Mean(1) umb 00083(2) Head 00068(3) Sternum 00054(4) Infraclavicular region 00060(5) Lumbar region 00025

Table 4 Seat marginal mean values of σa (in g)

Seat Mean(1) Zafu 00059(2) Chair 00056(3) Bench 00060

Table 5 ANOVA results for the anterior-posterior sway (Axz) andthe left-right sway (Axy) versus the seat

Axz versus seat Axy versus seatp value 0196 0002p values less than 005 are highlighted in bold

Table 6 Values obtained for each seat considering anterior-pos-terior and left-right sways (Axz and Axy in degrees)

Seat Axz AxyZafu 678 (751) 583 (511)Chair 848 (862) 1001 (972)Meditation bench 1013 (1292) 918 (687)Values are given as mean (SD)

Table 7 Seat analysis of the left-right sway (Axy) pairwisecomparisons with Bonferronirsquos correction

Zafu ChairChair 0019 mdashMeditation bench 0006 1000p values less than 005 are highlighted in bold

ndash10

ndash15

ndash20

ndash25

ndash300 05 1

Samples at 50 Hz times104

Deg

rees

ndash10

ndash15

ndash20

ndash25

ndash30

Deg

rees

15 2 25 3 0 05 1Samples at 50 Hz times104

15 2 25

Figure 5 Signal before the lter (a) and signal after the lter (b)

6 Journal of Healthcare Engineering

time Western teachers and mindfulness professionals arenot so strict about this point instead of a cushion (zafu)a comfortable chair or a meditation bench can be used too Apriori one could think that the chair should give someadvantage since most of the volunteers had no meditationexperience and therefore they were not used to sit on thezafu or the meditation bench which were probably un-known for them +e only significant difference was relatedto the global change in posture throughout the session (left-right sway) +e zafu provided the best seat with significantdifferences from the chair and the meditation bench inkeeping with the oriental tradition However anterior-posterior sway and σa were not significantly different+us it might seem reasonable to advise using the zafu +isis consistent with the research in Zen sitting postures +euse of the zafu to maintain fixed postures has been applied inother fields such as in the development of a chair for mi-croscopic surgery [51] or in the improvement of the sittingposture for children in the classrooms [52]

However other aspects such as comfort should beconsidered by western and beginner meditators In thisregard some people reported pain in the anterior part of theankle when sitting on a meditation bench People with backpain or difficulty getting up from a low seated positionshould consider a comfortable chair [17]

In the context of breathing interventions this study hasserved to give a tip on the seat for meditation used in thebeginning stages of training It is supported by stabilitymeasures +e determination of sensor location is importantto reduce the number of locations to be monitored in view offuture studies in which the relation between stability andbeneficial outcomes of mindfulness should be investigatedAnyway the results of this study are preliminary and shouldbe taken with caution

One of the limitations of this work is the number ofvolunteers which corresponds to a pilot study to check thewearable sensor kit +e location of the sensors also entailssome random components due to difficulties placing them inan exact position However these differences are very lowcompared to the distance between different locations An-other concern is the fact that the accelerometers used in thisstudy include parts from different manufacturers Howeverthe measurements were taken in almost static situations andtherefore their range was irrelevant Besides we performedthe analysis described in Section 222 to lower the resolutionof the sensors to a common baseline +e main conclusionsof the statistical analyses remained the same after thispreprocessing

5 Conclusions and Future Lines of Work

Several review articles support the usefulness and effec-tiveness of mindfulness techniques in health and well-being+us many people are starting to practice mindfulness Inthe beginning stages daily practice between 5 and 15minutes is recommended in which the stable posture is a keycomponent In this work we have presented a set of inertialsensors to measure the motion and change in posture +econventionally used lumbar region is not the best body

location for sensors since it presents lower sensitivity Be-sides out of all the meditation seats the zafu has someadvantages in one out of three parameters obtained in thisstudy concerning the overall change in posture from theleft-right sway

In summary a kit for measuring stability in meditationcould consider a single sensor placed at the head if the useris willing to wear an accessory like a pair of glasses or at-tached to clothes at the infraclavicular region An additionalsensor at the thumb finger could also be used in certainforms of meditation [50] since its movements are consideredby many meditation masters as a direct indication of themeditation depth In addition beginners could start prac-ticing meditation in chairs or meditation benches althougha transition to the zafu is recommended for a medium term

+e present study could be improved in several waysFurther improvements in the orientation angles can beobtained with more complex low-pass filters or estimationwith data fusion of the accelerometer gyroscope andmagnetometer using compensatory filters +us gravitycould be determinedmore properly With all these the studycould go beyond the assumptions of a single force in thesystem (gravity) or a fixed orientation of the glasses sensorAnother hardware issue is power consumption especially iflonger sessions were to be recorded It should be measuredand reduced for instance using low-power Bluetooth (forinstance the module CC2541) +e possibility of data lossesin the Bluetooth transmission should also be considered andhandled in a future smartphone application +e measure-ments should be extended with more volunteers improvingthis pilot research In a future study the sensors will be usedto provide feedback information +e effect of this feedbackonmindfulness training programs and self-regulation will beexplored Specifically changes in the proprioception (theability to sense the position location and movement of thebody and its parts) could be evaluated with the practiceMoreover the role of movement and posture in bringingabout beneficial outcomes of mindfulness-based interven-tions should be investigated

Appendix

Detailed Description of the Statistical Test

In this section a more detailed description of the statisticaltests is given For the ANOVA analysis of σa with repeatedmeasures (factors sensor location and seat) the results areshown in Table 8 Neither the sensor location factor nor theinteraction factor fulfilled the sphericity condition so that forthem the p value of Pillairsquos trace was selected Othercommon statistics provided by SPSS like the GreenhousendashGeisser correction led to the same conclusion as Pillairsquos trace

+e ANOVA results of the analysis of the sway (Axy andAxz) with respect to the seat are shown in Table 9 No mea-surement fulfilled the sphericity test so the value of Pillairsquos tracewas selected In the case ofAxy the null hypothesis was rejectedOther common statistics of SPSS (GreenhousendashGeisser cor-rection) led to the same conclusion

Journal of Healthcare Engineering 7

Data Availability

+e data used in this article are available upon request fromthe first author

Conflicts of Interest

+e authors declare no conflicts of interest

Authorsrsquo Contributions

Inmaculada Plaza and Carlos T Medrano conceived anddesigned the experiments Victor H Rodriguez developedthe measuring kit Victor H Rodriguez and InmaculadaPlaza performed the experiments Victor H Rodriguezanalyzed the data Carlos T Medrano contributed analysistools Inmaculada Plaza Victor H Rodriguez and CarlosT Medrano wrote the paper

Acknowledgments

+e authors would like to thank all volunteers who partici-pated in the study +anks are due to ldquoCONACYT-Gobiernodel Estado de Durango Mexicordquo for the research-grant ref-erence 330795386043+is workwas supported by the ldquoFondoSocial Europeordquo the ldquoGobierno de Aragonrdquo the Ministry ofEconomy and Competitiveness (Project reference TEC2013-50049-EXP) and the University of Zaragoza (Project referenceUZ2017-TEC-02)

References

[1] J Kabat-Zinn ldquoAn outpatient program in behavioral medi-cine for chronic pain patients based on the practice ofmindfulness meditation theoretical considerations and pre-liminary resultsrdquo General Hospital Psychiatry vol 4 no 1pp 33ndash47 1982

[2] J Kabat-Zinn ldquoMindfulness-based interventions in contextpast present and futurerdquo Clinical Psychology Science andPractice vol 10 pp 144ndash156 2003

[3] R A Baer ldquoMindfulness training as clinical interventiona conceptual and empirical reviewrdquo Clinical PsychologyScience and Practice vol 10 pp 125ndash143 2003

[4] P Grossman L Niemann S Schmidt and H WalachldquoMindfulness-based stress reduction and health benefits Ameta-analysisrdquo Journal of Psychosomatic Research vol 57 no 1pp 5ndash43 2004

[5] P Sedlmeier M Eberth D Schwarz et al ldquo+e psychologicaleffects of meditation a meta-analysisrdquo Psychological Bulletinvol 138 no 6 pp 1139ndash1171 2012

[6] M Goyal S Singh N Gould et al ldquoMeditation programs forpsychological stress and well-being a systematic review andmeta-analysisrdquo JAMA Internal Medicine vol 174 no 3pp 357ndash368 2014

[7] B Khoury B Knauper F Pagnini N Trent A Chiesa andK Carriere ldquoEmbodied mindfulnessrdquo Mindfulness vol 8no 5 pp 1ndash12 2017

[8] R Aloneftis ldquo+e role of embodiment in third wave mind-fulness based cognitive therapiesrdquo PsyPAG Quarterly vol 102pp 40ndash43 2017

[9] J M Burg T Probst T Heidenreich and J Michalak ldquoDe-velopment and psychometric evaluation of the bodymindfulnessquestionnairerdquo Mindfulness vol 8 no 3 pp 807ndash818 2017

[10] A Cebolla L Galiana D Campos et al ldquoHow does mind-fulness work Exploring a theoretical model using samples ofmeditators and non-meditatorsrdquoMindfulness pp 1ndash11 2017

[11] S Alidina Mindfulness for Dummies John Wiley amp SonsHoboken NJ USA 2014

[12] E ShoninW Van Gordon andM D Griffiths ldquoPractical tipsfor using mindfulness in general practicerdquo British Journal ofGeneral Practice vol 64 no 624 pp 368-369 2014

[13] S R Bishop M Lau S Shapiro et al ldquoMindfulness a pro-posed operational definitionrdquo Clinical Psychology Science andPractice vol 11 no 3 pp 230ndash241 2004

[14] J Kabat-Zinn ldquoSitting meditationsrdquoMindfulness vol 7 no 6pp 1441ndash1444 2016

[15] C E Jones An Investigation into the Role of Body Posture inMindfulness Practice Doctoral dissertation Canterbury ChristChurch University Canterbury UK 2016

[16] T N Hanh5eMiracle of Mindfulness Beacon Boston MAUSA 1975

[17] C P Cook-CottoneMindfulness and Yoga for Self-RegulationA Primer for Mental Health Professionals Springer PublishingCompany New York NY USA 1st edition 2015

[18] P Kaul J Passafiume R C Sargent and B F OrsquoHaraldquoMeditation acutely improves psychomotor vigilance andmay decrease sleep needrdquo Behavioral and Brain Functionsvol 6 no 1 p 1 2010

[19] A Ahani M Wahbeh M Miller et al ldquoChange in physio-logical signals during mindfulness meditationrdquo in Proceedingsof the IEEE EMBS Conference Osaka Japan July 2013

[20] J J Arch and M G Craske ldquoMechanisms of mindfulnessemotion regulation following a focused breathing inductionrdquoBehaviour Research and 5erapy vol 44 no 12 pp 1849ndash1858 2006

[21] J Vidyarthi and B E Riecke ldquoInteractively mediating ex-periences of mindfulness meditationrdquo International Journal ofHuman-Computer Studies vol 72 no 8 pp 674ndash688 2014

[22] W Y Wong and M S Wong ldquoTrunk posture monitoringwith inertial sensorsrdquo European Spine Journal vol 17 no 5pp 743ndash753 2008

[23] M Dozza L Chiari and F B Horak ldquoA portable audio-biofeedback system to improve postural controlrdquo in Pro-ceedings of the 26th Annual International Conference of the

Table 8 ANOVA results for σa (p values)

Mauchlyrsquos test of sphericitySeat 0239

Sensor location 0000Seatlowast sensor location 0000

Pillairsquos traceSeat 0675

Sensor location 0000Seatlowast sensor location 0 726

Sphericity assumedSeat 0744

Sensor location 0000Seatlowast sensor location 0725

Table 9 ANOVA results for the sway (Axz and Axy versus seat)(p value)

Axz Axy

Mauchlyrsquos test of sphericity 0000 0037Pillairsquos trace 0196 0002Sphericity assumed 0299 0006

8 Journal of Healthcare Engineering

IEEE Engineering in Medicine and Biology Society pp 4799ndash4802 San Francisco CA USA September 2004

[24] S L Whitney J L Roche G F Marchetti et al ldquoA com-parison of accelerometry and center of pressure measuresduring computerized dynamic posturography a measure ofbalancerdquo Gait and Posture vol 33 no 4 pp 594ndash599 2011

[25] W Maetzler M Mancini I Liepelt-Scarfone et al ldquoImpairedtrunk stability in individuals at high risk for Parkinsonrsquosdiseaserdquo PLoS One vol 7 no 3 article e32240 pp 1ndash6 2012

[26] J Houtan R Changbum T Ahn and L Stentz ldquoFall riskanalysis of construction workers using inertial measurementunits validating the usefulness of the postural stability metricsin constructionrdquo Safety Science vol 84 pp 161ndash170 2016

[27] S J Ozinga and J L Alberts ldquoQuantification of posturalstability in older adults using mobile technologyrdquo Experi-mental Brain Research vol 232 no 12 pp 3861ndash3972 2014

[28] S R Lord C Sherrington and H B Menz Falls in OlderPeople Risk Factors and Strategies for Prevention CambridgeUniversity Press Cambridge UK 2nd edition 2007

[29] K Chang S Chen and H Lee ldquoWireless accelerator basedbody posture stability detection and application for medita-tion practitionersrdquo in Proceedings of the International Con-ference on Biomedical and Health Informatics BHIpp 112ndash115 Hong Kong China January 2012

[30] K Chang S Chen H Lee et al ldquoA wireless accelerometer-based body posture stability detection system and its appli-cation for meditation practitionersrdquo Sensors vol 12 no 12pp 17620ndash17632 2012

[31] K M Chang Y T Chun S H Chen et al ldquo+e evaluation ofphysical stillness with wearable chest and arm accelerometerduring Chan Ding practicerdquo Sensors vol 16 p 1126 2016

[32] P M Deshmukh C M Russell L E Lucarino andS N Robinovitch ldquoEnhancing clinical measures of posturalstability with wearable sensorsrdquo in Proceedings of the 34th AnnualInternational Conference of the IEEE EMBS pp 4521ndash4524 SanDiego CA USA September 2012

[33] R P Hubble G A Naughton P A Silburn and M H ColeldquoWearable sensor use for assessing standing balance andwalking stability in people with Parkinsonrsquos disease a system-atic reviewrdquo PloS One vol 10 no 4 article e0123705 2015

[34] Lumo Web Page of Lumo Lift product 2017 httpswwwlumobodytechcomlumoliftutm_expid70259375-21Yz80XOgnSXu9lXiNTep1yA1amputm_referrerhttps3A2F2Fwwwlumobodytechcom2Fscience-of-lumo-run2F

[35] Upright Web Page of Personal Posture Trainer-UprightProduct 2017 httpwwwuprightposecom

[36] Alex Web Page of Alex Product 2017 httpalexposturecom[37] Jins-Meme Web Page of Jins Meme Glasses 2017 httpsjins-

memecomen[38] PGrossman ldquoMindfulness for psychologists paying kind attention

to the perceptiblerdquo Mindfulness vol 1 no 2 pp 87ndash97 2010[39] IMU9150 2017 httpswwwinvensensecomproductsmotion-

tracking9-axismpu-9150[40] ADXL345 2017 httpwwwanalogcomenproductsmems

accelerometersadxl345html[41] Arduino Pro Mini 2017 httpswwwarduinoccenMain

ArduinoBoardProMini[42] Bluetooth HC-05 2017 httpwwwthaieasyeleccomdownloads

EFDV390EFDV390_Datasheetpdf[43] B Najafi K Aminian A Paraschiv-Ionescu F Loew C J Bula

and P Robert ldquoAmbulatory system for human motion analysisusing a kinematic sensor monitoring of daily physical activityin the elderlyrdquo IEEE Transactions on Biomedical Engineeringvol 50 no 6 pp 711ndash723 2003

[44] K T Wyne ldquoA comprehensive review of tremorrdquo JAAPAvol 18 no 2 pp 43ndash50 2005

[45] S Gottlieb and O C J Lippold ldquo+e 4-6 Hz tremor duringsustained contraction in normal human subjectsrdquo Journal ofPhysiology vol 336 no 1 pp 499ndash509 1983

[46] M C Barroso G P Esteves T P Nunes L M G SilvaA C D Faria and P L Melo ldquoA telemedicine instrument forremote evaluation of tremor design and initial applications infatigue and patients with Parkinsonrsquos diseaserdquo BiomedicalEngineering OnLine vol 10 no 1 p 14 2011

[47] Android Developers Sensor Manager 2018 httpsdeveloperandroidcomreferenceandroidhardwareSensorManagerhtml

[48] T Deshimaru and Y Ikemi Zen et Self-Control Kairos S AEditions Retz Barcelona Spain 6th edition 1985

[49] K Sekida Zen Training Kairos S A Weatherhill Inc NewYork NY USA 6th edition 2016

[50] D Villalba What Is Zen Miraguano Ediciones MadridSpain 2005 ISBN84-7813-286-4

[51] K Noro T Naruse R Lueder N Nao-i and M KozawaldquoApplication of Zen sitting principles to microscopic surgeryseatingrdquoApplied Ergonomics vol 43 no 2 pp 308ndash319 2012

[52] K Ishihara K Dake T Kashihara and S Ishihara ldquoAn attemptto improve the sitting posture of children in the classroomsrdquo inProceedings of the International MultiConference of Engineersand Computer Scientists vol 3 Hong Kong March 2010

Journal of Healthcare Engineering 9

During the meditation sessions a series of inertial sensorswere placed on dierent body locations as described in Figure 2

More specically the sensors were placed as follows

(1) An IMU9150 [39] (plusmn4G 16-bit resolution) on thehand specically on the left thumb fastened withstrips (from now on this location will be referredto in short as thumb)

(2) An IMU9150 on the head held by glasses (inshort head)

(3) An ADXL345 [40] (plusmn4G 13-bit resolution) on thexiphoid process of the sternum (in short sternum)

(4) An IMU9150 on the left infraclavicular region onthe pectoralis major just below the collarbone (inshort infraclavicular region)

(5) A smartphone (Samsung Galaxy Trend Plus GT-S7580 accelerometer with plusmn2G 12-bit resolu-tion) on the left side of the lumbar inside a specialbag (in short lumbar region)

(a) (b) (c)

Figure 1 Meditation seats used in the short meditation sessions (a) chair (b) zafu (c) meditation bench

Head (held by glasses)(IMU9150)

Infraclavicular region(IMU9150)

Sternum(ADXL345)

Le side of the lumbar(smartphone (GT-S7580))

Le thumb(IMU9150)

Figure 2 Schematic location of sensors on the body

Journal of Healthcare Engineering 3

e sensors were selected for convenience availability inour laboratory and ease of use with libraries for the Arduinoplatform available although many commercial parts couldhave been selected too ey fulll the requirements of theapplication (Section 221) Since a smartphone was used togather data (as it would be in a real application of the system)we took advantage of its accelerometer to add another lo-cation (lumbar region) ere was no relation between thesensor and location other than the convenience of placing thesmartphone in the lumbar region with a commercial bag

221 Signal Acquisition Signals were acquired through anArduino Pro Mini [41] which gathered information from foursensors (inertial sensor) with a frequency of 50Hze librariesof the inertial sensors were provided by SparkFun makinguse of the I2C protocol for communication e IMU9150only has two addresses in the same I2C bus Since our kit hadthree IMU9150 a second I2C bus was required e ArduinoPro Mini has only one built-in I2C bus (hardware imple-mentation)erefore the second onewas emulated by softwareAll sensor information was sent in real time to a smartphonethrough an HC-05 Bluetooth interface [42] at a rate of 115200bauds is serial communication was the bottleneck in thetransmission Sensor data were transmitted as ASCII char-acters and the maximum transmission frequency with thatserial port speed was 68Hz given the number of charactersper sampling period in our experiment higher than the selectedfrequency While the smartphone collected informationfrom the Arduino it sensed meditator movement with itsinternal accelerometer at a frequency of 50 Hz

e frequency was selected according to the followingreasoning Basically the movement is expected to be underalmost static conditions thus a low-frequency measurementEven normal movements like gait are of low frequency (gait ismeasured in 06ndash5Hz in [43]) One could also consider thepossibility of tremor induced by the posture Enhancedphysiological tremor and even pathological tremor are below18Hz [44] and other studies have reported detecting tremorcaused by fatigue in a low frequency of 2ndash6Hz in [45] or 2Hzpeak in [46]us a sampling frequency of 50Hz was enoughand allowed a comfortable wireless Bluetooth transmissionMinimal data losses were reported in our experiments andthe corresponding values were found by interpolation in thepostprocessing Data losses could be due to transmissionerrors or due to errors when storing data in les Finallyexperimentally it seems that 50Hz corresponds to one of thesampling options in the selected smartphone model (optionldquoSENSOR_DELAY__GAMErdquo in the Android source code[47]) with some variability since this device cannot deal withstrict real time and the values have to be interpolated too

After every meditation session the information stored ina le was labeled with the name of the seat (chair zafu andmeditation bench) and the subjectrsquos identier

222 Signal Processing e les were processed withMATLAB e rst 60 and last 30 seconds of each sessionwere removed to avoid incorrect data coming from peoplesitting down or from the smartphone being removed from

the special bag Two types of processing were performedFirst the acceleration norm for each sensor was calculatedand then its standard deviation (σa) during each meditationsession was extracted is value is considered a conven-tional measure of stability that is sensitive to small and rapidchanges in the posture such as tremors due to not beingable to maintain the posture

Since parts of dierent manufacturers were used theprocessing was repeated by discarding the least signicantbits of the most accurate sensors In this way we simulateda situation in which all the sensors had the same resolution(the worst one) and we tested the inordfuence on the results

Subsequently the accelerometer signal was processed toobtain a global measure of changes in the volunteerrsquos positione idea was to distinguish major changes during the sessionsey could be caused by volunteers adopting a dierentposition due to their inability to hold the previous one or dueto slow posture variations For this purpose the study focusedon the sensor placed on the head Specically the sensor wasplaced on the temple of the glasses since this location allowsfor an easy interpretation of deviations from the uprightposition in terms of acceleration components e approxi-mate orientation of the sensor is as shown in Figure 3 It isreasonable to assume that the acceleration is mainly due togravity because the movements while meditating are slowBesides orientation values obtained from accelerometerreadings were subjected to a low-pass ltering process (see (2))to enhance their reliability As shown in Figure 3 the maincomponent of the acceleration is along the x-axis Tilts aroundthe left-right direction or around the anterior-posterior di-rection led to an increase of y and z components respectively

us the angles of the acceleration with respect to theplanes xz and xy were measured using the followingequations

Axz tanminus1aya2x + a2zradic

Axy tanminus1az

a2x + a2yradic

(1)

Figure 4 shows the two measured angles e Axz anglecorresponds to an anterior-posterior movement and theAxy angle corresponds to a left-right movement

Gravity force

x

y

Figure 3 Approximate orientation of the sensor with respect to theEarthrsquos gravity force

4 Journal of Healthcare Engineering

After recording the angles they were filtered usinga moving average filter (500 samples 10 s) to reduce noise((2) and Figure 5) which effectively removed any dynamiccomponent

y(n) 1N

1113944

iNminus1

i0x(nminus i) (2)

where N 500A global measure of the volunteerrsquos change of position

during the session was then obtained from the differencebetween the maximum and minimum values of the signalFor example the signal in Figure 5 has a variation of 1245deg

223 Statistical Analysis +e standard deviation of theacceleration was analyzed using a two-way ANOVA withrepeated measures on both factors the sensor location andthe seat (results in Section 31) Similarly to analyze theglobal change of the posture (angle differences) an ANOVAwith one factor (seat) was carried out since in this secondcase only one sensor location was used (Section 32) +estatistical analyses were performed with SPSS 120

3 Results

31 Analysis of Accelerometer Standard Deviation Twofactors have been considered in the analysis the meditationseat during the sessions (ldquoseatrdquo chair zafu or meditationbench) and the location of sensors on the body (ldquosensorlocationrdquo thumb head sternum infraclavicular region orlumbar region)

Table 1 shows the results of an ANOVA with repeatedmeasures on both factors +e details about the statisticaltests performed are given in Appendix +e sensor locationwas found to be significant (p valuelt 005) On the otherhand neither the seat nor the interaction term was signif-icant Since the sensor location was found to be significantwe proceeded with a multiple comparison analysis to dis-cover the root of the differences in terms of pairwisecomparisons of sensor locations +e p values of thecomparisons are shown in Table 2 after applying a standardBonferronirsquos correction

Table 3 shows the mean values obtained for each of thelocations of the sensors and Table 4 shows the mean valuesobtained for each of the seats (marginal values provided by

SPSS) +e marginal values are obtained by averaging overone of the factors keeping only the other one +ey can givea clue of the influence of a single factor According to Ta-bles 2 and 3 the lumbar position is the least sensitive since itsdifferences with the others are significant and themean valueis the lowest one Among the remaining positions (thumbhead sternum and infraclavicular region) the thumb showsthe highest sensitivity although the interdifferences are notsignificant except with the sternum+e marginal values forthe seat factor in Table 4 are presented for completeness andto show that the experimental values are very close

+e results of this statistical analysis remained almostthe same after preprocessing the accelerometer to lower theresolution to the worst one as explained in Section 222+e only difference is that sternum and thumb did not differsignificantly after that preprocessing

32 Posture Variation Depending on the Seat +e results ofthe ANOVA for the anterior-posterior sway (Axz) and theleft-right sway (Axy) are presented in Table 5 Table 6 showsthe values of the sway for the three seats

+e zafu seems to permit a better posture setting that canbe held throughout the session since sways were lowerHowever for Axz the effect was not significant while forAxy it was +erefore a post hoc analysis was performed forAxy +e results are presented in Table 7 +e pairwisecomparisons indicate that the zafu seat was different fromthe others

4 Discussion

Postural stability is an important factor during mindfulnessmeditation training Inertial sensors can be a good option formonitoring posture dynamics due to their low cost and easeof use One of the goals of this paper was to determine whichis the most sensitive sensor location Multiple protocols weredescribed in the literature for the placement of wearablesensors on the human body for assessing for instancestanding balance or walking stability Most of them reportedplacing a wearable sensor on either the lumbar or sacralregion of the trunk To evaluate body posture stability duringsitting meditation Chang et al [30] secured the measure-ment system between the abdomen and the chest or on thearm [31] Manufacturers of commercial products choseother body positions collarbone head lower back and soon +e lumbar region is often selected since it is close to thebody center of mass in a standing posture However theresults of the present study indicate that the lumbar or sacralregion is not the best option for measuring motion duringsitting meditation since its σa value is the lowest and sig-nificantly different from the rest It could be ruled out forfuture applications Among the remaining locations it is notpossible to separate them into different groups and only thethumb is significantly different from the sternum+ereforeconvenience or comfort can be important factors to selectthe location +e head location requires a more bulky ac-cessory +e infraclavicular region is less invasive since thesensor is attached to clothes +e sternum and lumbar could

Angle Axy

Angle Axy

Angle Axz

Angle Axz

Figure 4 Direction of rotation angles with respect to glasses Bluearrow anterior-posterior (Axz) white arrow left-right (Axy)

Journal of Healthcare Engineering 5

be more accepted by users since sensors attached to strapsare common in commercial products for tness Finally thethumb location would require a miniaturization of thesystem to improve comfortability Usability acceptance andcomfort should be evaluated among users Although thedierences are not statistically signicant the sensitivity of

the head location was higher than the sensitivity of thesternum or the infraclavicular locations is seems rea-sonable since the distance to the seat base is larger on thehead In fact the order of the sensitivity among the locationshead infraclavicular region sternum and lumbar regionfollowed the order of the distance to the seat e thumbshowed an even larger degree of movement but this could berelated to the movements of the hand that can be in-dependent of upper body displacements In fact handposture is very important according to traditional texts[48 49] and in particular thumb ngers should not drop orrise but remain in a perfect horizontal position [50]

e other goal of this paper was to analyze the inordfuenceof the seat to recommend a specic one based on dataobtained using inertial sensors In the oriental traditionmeditation is performed while seated on a cushion pref-erentially in the full-lotus or half-lotus position e staticposition is not to be changed until the end of the meditation

Table 1 ANOVA results for σa (p values)

Factor p valueSeat 0744Sensor location 0000Seatlowast sensor location 0726p values less than 005 are in bold

Table 2 Sensor location analysis pairwise comparisons withBonferronirsquos correction

umb Head Sternum Infraclavicularregion

Head 1000 mdash mdash mdashSternum 0022 0329 mdash mdashInfraclavicularregion 0787 1000 1000 mdash

Lumbar region 0000 0000 0000 0000p values less than 005 are highlighted in bold

Table 3 Sensor location marginal mean values of σa (in g)

Sensor location Mean(1) umb 00083(2) Head 00068(3) Sternum 00054(4) Infraclavicular region 00060(5) Lumbar region 00025

Table 4 Seat marginal mean values of σa (in g)

Seat Mean(1) Zafu 00059(2) Chair 00056(3) Bench 00060

Table 5 ANOVA results for the anterior-posterior sway (Axz) andthe left-right sway (Axy) versus the seat

Axz versus seat Axy versus seatp value 0196 0002p values less than 005 are highlighted in bold

Table 6 Values obtained for each seat considering anterior-pos-terior and left-right sways (Axz and Axy in degrees)

Seat Axz AxyZafu 678 (751) 583 (511)Chair 848 (862) 1001 (972)Meditation bench 1013 (1292) 918 (687)Values are given as mean (SD)

Table 7 Seat analysis of the left-right sway (Axy) pairwisecomparisons with Bonferronirsquos correction

Zafu ChairChair 0019 mdashMeditation bench 0006 1000p values less than 005 are highlighted in bold

ndash10

ndash15

ndash20

ndash25

ndash300 05 1

Samples at 50 Hz times104

Deg

rees

ndash10

ndash15

ndash20

ndash25

ndash30

Deg

rees

15 2 25 3 0 05 1Samples at 50 Hz times104

15 2 25

Figure 5 Signal before the lter (a) and signal after the lter (b)

6 Journal of Healthcare Engineering

time Western teachers and mindfulness professionals arenot so strict about this point instead of a cushion (zafu)a comfortable chair or a meditation bench can be used too Apriori one could think that the chair should give someadvantage since most of the volunteers had no meditationexperience and therefore they were not used to sit on thezafu or the meditation bench which were probably un-known for them +e only significant difference was relatedto the global change in posture throughout the session (left-right sway) +e zafu provided the best seat with significantdifferences from the chair and the meditation bench inkeeping with the oriental tradition However anterior-posterior sway and σa were not significantly different+us it might seem reasonable to advise using the zafu +isis consistent with the research in Zen sitting postures +euse of the zafu to maintain fixed postures has been applied inother fields such as in the development of a chair for mi-croscopic surgery [51] or in the improvement of the sittingposture for children in the classrooms [52]

However other aspects such as comfort should beconsidered by western and beginner meditators In thisregard some people reported pain in the anterior part of theankle when sitting on a meditation bench People with backpain or difficulty getting up from a low seated positionshould consider a comfortable chair [17]

In the context of breathing interventions this study hasserved to give a tip on the seat for meditation used in thebeginning stages of training It is supported by stabilitymeasures +e determination of sensor location is importantto reduce the number of locations to be monitored in view offuture studies in which the relation between stability andbeneficial outcomes of mindfulness should be investigatedAnyway the results of this study are preliminary and shouldbe taken with caution

One of the limitations of this work is the number ofvolunteers which corresponds to a pilot study to check thewearable sensor kit +e location of the sensors also entailssome random components due to difficulties placing them inan exact position However these differences are very lowcompared to the distance between different locations An-other concern is the fact that the accelerometers used in thisstudy include parts from different manufacturers Howeverthe measurements were taken in almost static situations andtherefore their range was irrelevant Besides we performedthe analysis described in Section 222 to lower the resolutionof the sensors to a common baseline +e main conclusionsof the statistical analyses remained the same after thispreprocessing

5 Conclusions and Future Lines of Work

Several review articles support the usefulness and effec-tiveness of mindfulness techniques in health and well-being+us many people are starting to practice mindfulness Inthe beginning stages daily practice between 5 and 15minutes is recommended in which the stable posture is a keycomponent In this work we have presented a set of inertialsensors to measure the motion and change in posture +econventionally used lumbar region is not the best body

location for sensors since it presents lower sensitivity Be-sides out of all the meditation seats the zafu has someadvantages in one out of three parameters obtained in thisstudy concerning the overall change in posture from theleft-right sway

In summary a kit for measuring stability in meditationcould consider a single sensor placed at the head if the useris willing to wear an accessory like a pair of glasses or at-tached to clothes at the infraclavicular region An additionalsensor at the thumb finger could also be used in certainforms of meditation [50] since its movements are consideredby many meditation masters as a direct indication of themeditation depth In addition beginners could start prac-ticing meditation in chairs or meditation benches althougha transition to the zafu is recommended for a medium term

+e present study could be improved in several waysFurther improvements in the orientation angles can beobtained with more complex low-pass filters or estimationwith data fusion of the accelerometer gyroscope andmagnetometer using compensatory filters +us gravitycould be determinedmore properly With all these the studycould go beyond the assumptions of a single force in thesystem (gravity) or a fixed orientation of the glasses sensorAnother hardware issue is power consumption especially iflonger sessions were to be recorded It should be measuredand reduced for instance using low-power Bluetooth (forinstance the module CC2541) +e possibility of data lossesin the Bluetooth transmission should also be considered andhandled in a future smartphone application +e measure-ments should be extended with more volunteers improvingthis pilot research In a future study the sensors will be usedto provide feedback information +e effect of this feedbackonmindfulness training programs and self-regulation will beexplored Specifically changes in the proprioception (theability to sense the position location and movement of thebody and its parts) could be evaluated with the practiceMoreover the role of movement and posture in bringingabout beneficial outcomes of mindfulness-based interven-tions should be investigated

Appendix

Detailed Description of the Statistical Test

In this section a more detailed description of the statisticaltests is given For the ANOVA analysis of σa with repeatedmeasures (factors sensor location and seat) the results areshown in Table 8 Neither the sensor location factor nor theinteraction factor fulfilled the sphericity condition so that forthem the p value of Pillairsquos trace was selected Othercommon statistics provided by SPSS like the GreenhousendashGeisser correction led to the same conclusion as Pillairsquos trace

+e ANOVA results of the analysis of the sway (Axy andAxz) with respect to the seat are shown in Table 9 No mea-surement fulfilled the sphericity test so the value of Pillairsquos tracewas selected In the case ofAxy the null hypothesis was rejectedOther common statistics of SPSS (GreenhousendashGeisser cor-rection) led to the same conclusion

Journal of Healthcare Engineering 7

Data Availability

+e data used in this article are available upon request fromthe first author

Conflicts of Interest

+e authors declare no conflicts of interest

Authorsrsquo Contributions

Inmaculada Plaza and Carlos T Medrano conceived anddesigned the experiments Victor H Rodriguez developedthe measuring kit Victor H Rodriguez and InmaculadaPlaza performed the experiments Victor H Rodriguezanalyzed the data Carlos T Medrano contributed analysistools Inmaculada Plaza Victor H Rodriguez and CarlosT Medrano wrote the paper

Acknowledgments

+e authors would like to thank all volunteers who partici-pated in the study +anks are due to ldquoCONACYT-Gobiernodel Estado de Durango Mexicordquo for the research-grant ref-erence 330795386043+is workwas supported by the ldquoFondoSocial Europeordquo the ldquoGobierno de Aragonrdquo the Ministry ofEconomy and Competitiveness (Project reference TEC2013-50049-EXP) and the University of Zaragoza (Project referenceUZ2017-TEC-02)

References

[1] J Kabat-Zinn ldquoAn outpatient program in behavioral medi-cine for chronic pain patients based on the practice ofmindfulness meditation theoretical considerations and pre-liminary resultsrdquo General Hospital Psychiatry vol 4 no 1pp 33ndash47 1982

[2] J Kabat-Zinn ldquoMindfulness-based interventions in contextpast present and futurerdquo Clinical Psychology Science andPractice vol 10 pp 144ndash156 2003

[3] R A Baer ldquoMindfulness training as clinical interventiona conceptual and empirical reviewrdquo Clinical PsychologyScience and Practice vol 10 pp 125ndash143 2003

[4] P Grossman L Niemann S Schmidt and H WalachldquoMindfulness-based stress reduction and health benefits Ameta-analysisrdquo Journal of Psychosomatic Research vol 57 no 1pp 5ndash43 2004

[5] P Sedlmeier M Eberth D Schwarz et al ldquo+e psychologicaleffects of meditation a meta-analysisrdquo Psychological Bulletinvol 138 no 6 pp 1139ndash1171 2012

[6] M Goyal S Singh N Gould et al ldquoMeditation programs forpsychological stress and well-being a systematic review andmeta-analysisrdquo JAMA Internal Medicine vol 174 no 3pp 357ndash368 2014

[7] B Khoury B Knauper F Pagnini N Trent A Chiesa andK Carriere ldquoEmbodied mindfulnessrdquo Mindfulness vol 8no 5 pp 1ndash12 2017

[8] R Aloneftis ldquo+e role of embodiment in third wave mind-fulness based cognitive therapiesrdquo PsyPAG Quarterly vol 102pp 40ndash43 2017

[9] J M Burg T Probst T Heidenreich and J Michalak ldquoDe-velopment and psychometric evaluation of the bodymindfulnessquestionnairerdquo Mindfulness vol 8 no 3 pp 807ndash818 2017

[10] A Cebolla L Galiana D Campos et al ldquoHow does mind-fulness work Exploring a theoretical model using samples ofmeditators and non-meditatorsrdquoMindfulness pp 1ndash11 2017

[11] S Alidina Mindfulness for Dummies John Wiley amp SonsHoboken NJ USA 2014

[12] E ShoninW Van Gordon andM D Griffiths ldquoPractical tipsfor using mindfulness in general practicerdquo British Journal ofGeneral Practice vol 64 no 624 pp 368-369 2014

[13] S R Bishop M Lau S Shapiro et al ldquoMindfulness a pro-posed operational definitionrdquo Clinical Psychology Science andPractice vol 11 no 3 pp 230ndash241 2004

[14] J Kabat-Zinn ldquoSitting meditationsrdquoMindfulness vol 7 no 6pp 1441ndash1444 2016

[15] C E Jones An Investigation into the Role of Body Posture inMindfulness Practice Doctoral dissertation Canterbury ChristChurch University Canterbury UK 2016

[16] T N Hanh5eMiracle of Mindfulness Beacon Boston MAUSA 1975

[17] C P Cook-CottoneMindfulness and Yoga for Self-RegulationA Primer for Mental Health Professionals Springer PublishingCompany New York NY USA 1st edition 2015

[18] P Kaul J Passafiume R C Sargent and B F OrsquoHaraldquoMeditation acutely improves psychomotor vigilance andmay decrease sleep needrdquo Behavioral and Brain Functionsvol 6 no 1 p 1 2010

[19] A Ahani M Wahbeh M Miller et al ldquoChange in physio-logical signals during mindfulness meditationrdquo in Proceedingsof the IEEE EMBS Conference Osaka Japan July 2013

[20] J J Arch and M G Craske ldquoMechanisms of mindfulnessemotion regulation following a focused breathing inductionrdquoBehaviour Research and 5erapy vol 44 no 12 pp 1849ndash1858 2006

[21] J Vidyarthi and B E Riecke ldquoInteractively mediating ex-periences of mindfulness meditationrdquo International Journal ofHuman-Computer Studies vol 72 no 8 pp 674ndash688 2014

[22] W Y Wong and M S Wong ldquoTrunk posture monitoringwith inertial sensorsrdquo European Spine Journal vol 17 no 5pp 743ndash753 2008

[23] M Dozza L Chiari and F B Horak ldquoA portable audio-biofeedback system to improve postural controlrdquo in Pro-ceedings of the 26th Annual International Conference of the

Table 8 ANOVA results for σa (p values)

Mauchlyrsquos test of sphericitySeat 0239

Sensor location 0000Seatlowast sensor location 0000

Pillairsquos traceSeat 0675

Sensor location 0000Seatlowast sensor location 0 726

Sphericity assumedSeat 0744

Sensor location 0000Seatlowast sensor location 0725

Table 9 ANOVA results for the sway (Axz and Axy versus seat)(p value)

Axz Axy

Mauchlyrsquos test of sphericity 0000 0037Pillairsquos trace 0196 0002Sphericity assumed 0299 0006

8 Journal of Healthcare Engineering

IEEE Engineering in Medicine and Biology Society pp 4799ndash4802 San Francisco CA USA September 2004

[24] S L Whitney J L Roche G F Marchetti et al ldquoA com-parison of accelerometry and center of pressure measuresduring computerized dynamic posturography a measure ofbalancerdquo Gait and Posture vol 33 no 4 pp 594ndash599 2011

[25] W Maetzler M Mancini I Liepelt-Scarfone et al ldquoImpairedtrunk stability in individuals at high risk for Parkinsonrsquosdiseaserdquo PLoS One vol 7 no 3 article e32240 pp 1ndash6 2012

[26] J Houtan R Changbum T Ahn and L Stentz ldquoFall riskanalysis of construction workers using inertial measurementunits validating the usefulness of the postural stability metricsin constructionrdquo Safety Science vol 84 pp 161ndash170 2016

[27] S J Ozinga and J L Alberts ldquoQuantification of posturalstability in older adults using mobile technologyrdquo Experi-mental Brain Research vol 232 no 12 pp 3861ndash3972 2014

[28] S R Lord C Sherrington and H B Menz Falls in OlderPeople Risk Factors and Strategies for Prevention CambridgeUniversity Press Cambridge UK 2nd edition 2007

[29] K Chang S Chen and H Lee ldquoWireless accelerator basedbody posture stability detection and application for medita-tion practitionersrdquo in Proceedings of the International Con-ference on Biomedical and Health Informatics BHIpp 112ndash115 Hong Kong China January 2012

[30] K Chang S Chen H Lee et al ldquoA wireless accelerometer-based body posture stability detection system and its appli-cation for meditation practitionersrdquo Sensors vol 12 no 12pp 17620ndash17632 2012

[31] K M Chang Y T Chun S H Chen et al ldquo+e evaluation ofphysical stillness with wearable chest and arm accelerometerduring Chan Ding practicerdquo Sensors vol 16 p 1126 2016

[32] P M Deshmukh C M Russell L E Lucarino andS N Robinovitch ldquoEnhancing clinical measures of posturalstability with wearable sensorsrdquo in Proceedings of the 34th AnnualInternational Conference of the IEEE EMBS pp 4521ndash4524 SanDiego CA USA September 2012

[33] R P Hubble G A Naughton P A Silburn and M H ColeldquoWearable sensor use for assessing standing balance andwalking stability in people with Parkinsonrsquos disease a system-atic reviewrdquo PloS One vol 10 no 4 article e0123705 2015

[34] Lumo Web Page of Lumo Lift product 2017 httpswwwlumobodytechcomlumoliftutm_expid70259375-21Yz80XOgnSXu9lXiNTep1yA1amputm_referrerhttps3A2F2Fwwwlumobodytechcom2Fscience-of-lumo-run2F

[35] Upright Web Page of Personal Posture Trainer-UprightProduct 2017 httpwwwuprightposecom

[36] Alex Web Page of Alex Product 2017 httpalexposturecom[37] Jins-Meme Web Page of Jins Meme Glasses 2017 httpsjins-

memecomen[38] PGrossman ldquoMindfulness for psychologists paying kind attention

to the perceptiblerdquo Mindfulness vol 1 no 2 pp 87ndash97 2010[39] IMU9150 2017 httpswwwinvensensecomproductsmotion-

tracking9-axismpu-9150[40] ADXL345 2017 httpwwwanalogcomenproductsmems

accelerometersadxl345html[41] Arduino Pro Mini 2017 httpswwwarduinoccenMain

ArduinoBoardProMini[42] Bluetooth HC-05 2017 httpwwwthaieasyeleccomdownloads

EFDV390EFDV390_Datasheetpdf[43] B Najafi K Aminian A Paraschiv-Ionescu F Loew C J Bula

and P Robert ldquoAmbulatory system for human motion analysisusing a kinematic sensor monitoring of daily physical activityin the elderlyrdquo IEEE Transactions on Biomedical Engineeringvol 50 no 6 pp 711ndash723 2003

[44] K T Wyne ldquoA comprehensive review of tremorrdquo JAAPAvol 18 no 2 pp 43ndash50 2005

[45] S Gottlieb and O C J Lippold ldquo+e 4-6 Hz tremor duringsustained contraction in normal human subjectsrdquo Journal ofPhysiology vol 336 no 1 pp 499ndash509 1983

[46] M C Barroso G P Esteves T P Nunes L M G SilvaA C D Faria and P L Melo ldquoA telemedicine instrument forremote evaluation of tremor design and initial applications infatigue and patients with Parkinsonrsquos diseaserdquo BiomedicalEngineering OnLine vol 10 no 1 p 14 2011

[47] Android Developers Sensor Manager 2018 httpsdeveloperandroidcomreferenceandroidhardwareSensorManagerhtml

[48] T Deshimaru and Y Ikemi Zen et Self-Control Kairos S AEditions Retz Barcelona Spain 6th edition 1985

[49] K Sekida Zen Training Kairos S A Weatherhill Inc NewYork NY USA 6th edition 2016

[50] D Villalba What Is Zen Miraguano Ediciones MadridSpain 2005 ISBN84-7813-286-4

[51] K Noro T Naruse R Lueder N Nao-i and M KozawaldquoApplication of Zen sitting principles to microscopic surgeryseatingrdquoApplied Ergonomics vol 43 no 2 pp 308ndash319 2012

[52] K Ishihara K Dake T Kashihara and S Ishihara ldquoAn attemptto improve the sitting posture of children in the classroomsrdquo inProceedings of the International MultiConference of Engineersand Computer Scientists vol 3 Hong Kong March 2010

Journal of Healthcare Engineering 9

e sensors were selected for convenience availability inour laboratory and ease of use with libraries for the Arduinoplatform available although many commercial parts couldhave been selected too ey fulll the requirements of theapplication (Section 221) Since a smartphone was used togather data (as it would be in a real application of the system)we took advantage of its accelerometer to add another lo-cation (lumbar region) ere was no relation between thesensor and location other than the convenience of placing thesmartphone in the lumbar region with a commercial bag

221 Signal Acquisition Signals were acquired through anArduino Pro Mini [41] which gathered information from foursensors (inertial sensor) with a frequency of 50Hze librariesof the inertial sensors were provided by SparkFun makinguse of the I2C protocol for communication e IMU9150only has two addresses in the same I2C bus Since our kit hadthree IMU9150 a second I2C bus was required e ArduinoPro Mini has only one built-in I2C bus (hardware imple-mentation)erefore the second onewas emulated by softwareAll sensor information was sent in real time to a smartphonethrough an HC-05 Bluetooth interface [42] at a rate of 115200bauds is serial communication was the bottleneck in thetransmission Sensor data were transmitted as ASCII char-acters and the maximum transmission frequency with thatserial port speed was 68Hz given the number of charactersper sampling period in our experiment higher than the selectedfrequency While the smartphone collected informationfrom the Arduino it sensed meditator movement with itsinternal accelerometer at a frequency of 50 Hz

e frequency was selected according to the followingreasoning Basically the movement is expected to be underalmost static conditions thus a low-frequency measurementEven normal movements like gait are of low frequency (gait ismeasured in 06ndash5Hz in [43]) One could also consider thepossibility of tremor induced by the posture Enhancedphysiological tremor and even pathological tremor are below18Hz [44] and other studies have reported detecting tremorcaused by fatigue in a low frequency of 2ndash6Hz in [45] or 2Hzpeak in [46]us a sampling frequency of 50Hz was enoughand allowed a comfortable wireless Bluetooth transmissionMinimal data losses were reported in our experiments andthe corresponding values were found by interpolation in thepostprocessing Data losses could be due to transmissionerrors or due to errors when storing data in les Finallyexperimentally it seems that 50Hz corresponds to one of thesampling options in the selected smartphone model (optionldquoSENSOR_DELAY__GAMErdquo in the Android source code[47]) with some variability since this device cannot deal withstrict real time and the values have to be interpolated too

After every meditation session the information stored ina le was labeled with the name of the seat (chair zafu andmeditation bench) and the subjectrsquos identier

222 Signal Processing e les were processed withMATLAB e rst 60 and last 30 seconds of each sessionwere removed to avoid incorrect data coming from peoplesitting down or from the smartphone being removed from

the special bag Two types of processing were performedFirst the acceleration norm for each sensor was calculatedand then its standard deviation (σa) during each meditationsession was extracted is value is considered a conven-tional measure of stability that is sensitive to small and rapidchanges in the posture such as tremors due to not beingable to maintain the posture

Since parts of dierent manufacturers were used theprocessing was repeated by discarding the least signicantbits of the most accurate sensors In this way we simulateda situation in which all the sensors had the same resolution(the worst one) and we tested the inordfuence on the results

Subsequently the accelerometer signal was processed toobtain a global measure of changes in the volunteerrsquos positione idea was to distinguish major changes during the sessionsey could be caused by volunteers adopting a dierentposition due to their inability to hold the previous one or dueto slow posture variations For this purpose the study focusedon the sensor placed on the head Specically the sensor wasplaced on the temple of the glasses since this location allowsfor an easy interpretation of deviations from the uprightposition in terms of acceleration components e approxi-mate orientation of the sensor is as shown in Figure 3 It isreasonable to assume that the acceleration is mainly due togravity because the movements while meditating are slowBesides orientation values obtained from accelerometerreadings were subjected to a low-pass ltering process (see (2))to enhance their reliability As shown in Figure 3 the maincomponent of the acceleration is along the x-axis Tilts aroundthe left-right direction or around the anterior-posterior di-rection led to an increase of y and z components respectively

us the angles of the acceleration with respect to theplanes xz and xy were measured using the followingequations

Axz tanminus1aya2x + a2zradic

Axy tanminus1az

a2x + a2yradic

(1)

Figure 4 shows the two measured angles e Axz anglecorresponds to an anterior-posterior movement and theAxy angle corresponds to a left-right movement

Gravity force

x

y

Figure 3 Approximate orientation of the sensor with respect to theEarthrsquos gravity force

4 Journal of Healthcare Engineering

After recording the angles they were filtered usinga moving average filter (500 samples 10 s) to reduce noise((2) and Figure 5) which effectively removed any dynamiccomponent

y(n) 1N

1113944

iNminus1

i0x(nminus i) (2)

where N 500A global measure of the volunteerrsquos change of position

during the session was then obtained from the differencebetween the maximum and minimum values of the signalFor example the signal in Figure 5 has a variation of 1245deg

223 Statistical Analysis +e standard deviation of theacceleration was analyzed using a two-way ANOVA withrepeated measures on both factors the sensor location andthe seat (results in Section 31) Similarly to analyze theglobal change of the posture (angle differences) an ANOVAwith one factor (seat) was carried out since in this secondcase only one sensor location was used (Section 32) +estatistical analyses were performed with SPSS 120

3 Results

31 Analysis of Accelerometer Standard Deviation Twofactors have been considered in the analysis the meditationseat during the sessions (ldquoseatrdquo chair zafu or meditationbench) and the location of sensors on the body (ldquosensorlocationrdquo thumb head sternum infraclavicular region orlumbar region)

Table 1 shows the results of an ANOVA with repeatedmeasures on both factors +e details about the statisticaltests performed are given in Appendix +e sensor locationwas found to be significant (p valuelt 005) On the otherhand neither the seat nor the interaction term was signif-icant Since the sensor location was found to be significantwe proceeded with a multiple comparison analysis to dis-cover the root of the differences in terms of pairwisecomparisons of sensor locations +e p values of thecomparisons are shown in Table 2 after applying a standardBonferronirsquos correction

Table 3 shows the mean values obtained for each of thelocations of the sensors and Table 4 shows the mean valuesobtained for each of the seats (marginal values provided by

SPSS) +e marginal values are obtained by averaging overone of the factors keeping only the other one +ey can givea clue of the influence of a single factor According to Ta-bles 2 and 3 the lumbar position is the least sensitive since itsdifferences with the others are significant and themean valueis the lowest one Among the remaining positions (thumbhead sternum and infraclavicular region) the thumb showsthe highest sensitivity although the interdifferences are notsignificant except with the sternum+e marginal values forthe seat factor in Table 4 are presented for completeness andto show that the experimental values are very close

+e results of this statistical analysis remained almostthe same after preprocessing the accelerometer to lower theresolution to the worst one as explained in Section 222+e only difference is that sternum and thumb did not differsignificantly after that preprocessing

32 Posture Variation Depending on the Seat +e results ofthe ANOVA for the anterior-posterior sway (Axz) and theleft-right sway (Axy) are presented in Table 5 Table 6 showsthe values of the sway for the three seats

+e zafu seems to permit a better posture setting that canbe held throughout the session since sways were lowerHowever for Axz the effect was not significant while forAxy it was +erefore a post hoc analysis was performed forAxy +e results are presented in Table 7 +e pairwisecomparisons indicate that the zafu seat was different fromthe others

4 Discussion

Postural stability is an important factor during mindfulnessmeditation training Inertial sensors can be a good option formonitoring posture dynamics due to their low cost and easeof use One of the goals of this paper was to determine whichis the most sensitive sensor location Multiple protocols weredescribed in the literature for the placement of wearablesensors on the human body for assessing for instancestanding balance or walking stability Most of them reportedplacing a wearable sensor on either the lumbar or sacralregion of the trunk To evaluate body posture stability duringsitting meditation Chang et al [30] secured the measure-ment system between the abdomen and the chest or on thearm [31] Manufacturers of commercial products choseother body positions collarbone head lower back and soon +e lumbar region is often selected since it is close to thebody center of mass in a standing posture However theresults of the present study indicate that the lumbar or sacralregion is not the best option for measuring motion duringsitting meditation since its σa value is the lowest and sig-nificantly different from the rest It could be ruled out forfuture applications Among the remaining locations it is notpossible to separate them into different groups and only thethumb is significantly different from the sternum+ereforeconvenience or comfort can be important factors to selectthe location +e head location requires a more bulky ac-cessory +e infraclavicular region is less invasive since thesensor is attached to clothes +e sternum and lumbar could

Angle Axy

Angle Axy

Angle Axz

Angle Axz

Figure 4 Direction of rotation angles with respect to glasses Bluearrow anterior-posterior (Axz) white arrow left-right (Axy)

Journal of Healthcare Engineering 5

be more accepted by users since sensors attached to strapsare common in commercial products for tness Finally thethumb location would require a miniaturization of thesystem to improve comfortability Usability acceptance andcomfort should be evaluated among users Although thedierences are not statistically signicant the sensitivity of

the head location was higher than the sensitivity of thesternum or the infraclavicular locations is seems rea-sonable since the distance to the seat base is larger on thehead In fact the order of the sensitivity among the locationshead infraclavicular region sternum and lumbar regionfollowed the order of the distance to the seat e thumbshowed an even larger degree of movement but this could berelated to the movements of the hand that can be in-dependent of upper body displacements In fact handposture is very important according to traditional texts[48 49] and in particular thumb ngers should not drop orrise but remain in a perfect horizontal position [50]

e other goal of this paper was to analyze the inordfuenceof the seat to recommend a specic one based on dataobtained using inertial sensors In the oriental traditionmeditation is performed while seated on a cushion pref-erentially in the full-lotus or half-lotus position e staticposition is not to be changed until the end of the meditation

Table 1 ANOVA results for σa (p values)

Factor p valueSeat 0744Sensor location 0000Seatlowast sensor location 0726p values less than 005 are in bold

Table 2 Sensor location analysis pairwise comparisons withBonferronirsquos correction

umb Head Sternum Infraclavicularregion

Head 1000 mdash mdash mdashSternum 0022 0329 mdash mdashInfraclavicularregion 0787 1000 1000 mdash

Lumbar region 0000 0000 0000 0000p values less than 005 are highlighted in bold

Table 3 Sensor location marginal mean values of σa (in g)

Sensor location Mean(1) umb 00083(2) Head 00068(3) Sternum 00054(4) Infraclavicular region 00060(5) Lumbar region 00025

Table 4 Seat marginal mean values of σa (in g)

Seat Mean(1) Zafu 00059(2) Chair 00056(3) Bench 00060

Table 5 ANOVA results for the anterior-posterior sway (Axz) andthe left-right sway (Axy) versus the seat

Axz versus seat Axy versus seatp value 0196 0002p values less than 005 are highlighted in bold

Table 6 Values obtained for each seat considering anterior-pos-terior and left-right sways (Axz and Axy in degrees)

Seat Axz AxyZafu 678 (751) 583 (511)Chair 848 (862) 1001 (972)Meditation bench 1013 (1292) 918 (687)Values are given as mean (SD)

Table 7 Seat analysis of the left-right sway (Axy) pairwisecomparisons with Bonferronirsquos correction

Zafu ChairChair 0019 mdashMeditation bench 0006 1000p values less than 005 are highlighted in bold

ndash10

ndash15

ndash20

ndash25

ndash300 05 1

Samples at 50 Hz times104

Deg

rees

ndash10

ndash15

ndash20

ndash25

ndash30

Deg

rees

15 2 25 3 0 05 1Samples at 50 Hz times104

15 2 25

Figure 5 Signal before the lter (a) and signal after the lter (b)

6 Journal of Healthcare Engineering

time Western teachers and mindfulness professionals arenot so strict about this point instead of a cushion (zafu)a comfortable chair or a meditation bench can be used too Apriori one could think that the chair should give someadvantage since most of the volunteers had no meditationexperience and therefore they were not used to sit on thezafu or the meditation bench which were probably un-known for them +e only significant difference was relatedto the global change in posture throughout the session (left-right sway) +e zafu provided the best seat with significantdifferences from the chair and the meditation bench inkeeping with the oriental tradition However anterior-posterior sway and σa were not significantly different+us it might seem reasonable to advise using the zafu +isis consistent with the research in Zen sitting postures +euse of the zafu to maintain fixed postures has been applied inother fields such as in the development of a chair for mi-croscopic surgery [51] or in the improvement of the sittingposture for children in the classrooms [52]

However other aspects such as comfort should beconsidered by western and beginner meditators In thisregard some people reported pain in the anterior part of theankle when sitting on a meditation bench People with backpain or difficulty getting up from a low seated positionshould consider a comfortable chair [17]

In the context of breathing interventions this study hasserved to give a tip on the seat for meditation used in thebeginning stages of training It is supported by stabilitymeasures +e determination of sensor location is importantto reduce the number of locations to be monitored in view offuture studies in which the relation between stability andbeneficial outcomes of mindfulness should be investigatedAnyway the results of this study are preliminary and shouldbe taken with caution

One of the limitations of this work is the number ofvolunteers which corresponds to a pilot study to check thewearable sensor kit +e location of the sensors also entailssome random components due to difficulties placing them inan exact position However these differences are very lowcompared to the distance between different locations An-other concern is the fact that the accelerometers used in thisstudy include parts from different manufacturers Howeverthe measurements were taken in almost static situations andtherefore their range was irrelevant Besides we performedthe analysis described in Section 222 to lower the resolutionof the sensors to a common baseline +e main conclusionsof the statistical analyses remained the same after thispreprocessing

5 Conclusions and Future Lines of Work

Several review articles support the usefulness and effec-tiveness of mindfulness techniques in health and well-being+us many people are starting to practice mindfulness Inthe beginning stages daily practice between 5 and 15minutes is recommended in which the stable posture is a keycomponent In this work we have presented a set of inertialsensors to measure the motion and change in posture +econventionally used lumbar region is not the best body

location for sensors since it presents lower sensitivity Be-sides out of all the meditation seats the zafu has someadvantages in one out of three parameters obtained in thisstudy concerning the overall change in posture from theleft-right sway

In summary a kit for measuring stability in meditationcould consider a single sensor placed at the head if the useris willing to wear an accessory like a pair of glasses or at-tached to clothes at the infraclavicular region An additionalsensor at the thumb finger could also be used in certainforms of meditation [50] since its movements are consideredby many meditation masters as a direct indication of themeditation depth In addition beginners could start prac-ticing meditation in chairs or meditation benches althougha transition to the zafu is recommended for a medium term

+e present study could be improved in several waysFurther improvements in the orientation angles can beobtained with more complex low-pass filters or estimationwith data fusion of the accelerometer gyroscope andmagnetometer using compensatory filters +us gravitycould be determinedmore properly With all these the studycould go beyond the assumptions of a single force in thesystem (gravity) or a fixed orientation of the glasses sensorAnother hardware issue is power consumption especially iflonger sessions were to be recorded It should be measuredand reduced for instance using low-power Bluetooth (forinstance the module CC2541) +e possibility of data lossesin the Bluetooth transmission should also be considered andhandled in a future smartphone application +e measure-ments should be extended with more volunteers improvingthis pilot research In a future study the sensors will be usedto provide feedback information +e effect of this feedbackonmindfulness training programs and self-regulation will beexplored Specifically changes in the proprioception (theability to sense the position location and movement of thebody and its parts) could be evaluated with the practiceMoreover the role of movement and posture in bringingabout beneficial outcomes of mindfulness-based interven-tions should be investigated

Appendix

Detailed Description of the Statistical Test

In this section a more detailed description of the statisticaltests is given For the ANOVA analysis of σa with repeatedmeasures (factors sensor location and seat) the results areshown in Table 8 Neither the sensor location factor nor theinteraction factor fulfilled the sphericity condition so that forthem the p value of Pillairsquos trace was selected Othercommon statistics provided by SPSS like the GreenhousendashGeisser correction led to the same conclusion as Pillairsquos trace

+e ANOVA results of the analysis of the sway (Axy andAxz) with respect to the seat are shown in Table 9 No mea-surement fulfilled the sphericity test so the value of Pillairsquos tracewas selected In the case ofAxy the null hypothesis was rejectedOther common statistics of SPSS (GreenhousendashGeisser cor-rection) led to the same conclusion

Journal of Healthcare Engineering 7

Data Availability

+e data used in this article are available upon request fromthe first author

Conflicts of Interest

+e authors declare no conflicts of interest

Authorsrsquo Contributions

Inmaculada Plaza and Carlos T Medrano conceived anddesigned the experiments Victor H Rodriguez developedthe measuring kit Victor H Rodriguez and InmaculadaPlaza performed the experiments Victor H Rodriguezanalyzed the data Carlos T Medrano contributed analysistools Inmaculada Plaza Victor H Rodriguez and CarlosT Medrano wrote the paper

Acknowledgments

+e authors would like to thank all volunteers who partici-pated in the study +anks are due to ldquoCONACYT-Gobiernodel Estado de Durango Mexicordquo for the research-grant ref-erence 330795386043+is workwas supported by the ldquoFondoSocial Europeordquo the ldquoGobierno de Aragonrdquo the Ministry ofEconomy and Competitiveness (Project reference TEC2013-50049-EXP) and the University of Zaragoza (Project referenceUZ2017-TEC-02)

References

[1] J Kabat-Zinn ldquoAn outpatient program in behavioral medi-cine for chronic pain patients based on the practice ofmindfulness meditation theoretical considerations and pre-liminary resultsrdquo General Hospital Psychiatry vol 4 no 1pp 33ndash47 1982

[2] J Kabat-Zinn ldquoMindfulness-based interventions in contextpast present and futurerdquo Clinical Psychology Science andPractice vol 10 pp 144ndash156 2003

[3] R A Baer ldquoMindfulness training as clinical interventiona conceptual and empirical reviewrdquo Clinical PsychologyScience and Practice vol 10 pp 125ndash143 2003

[4] P Grossman L Niemann S Schmidt and H WalachldquoMindfulness-based stress reduction and health benefits Ameta-analysisrdquo Journal of Psychosomatic Research vol 57 no 1pp 5ndash43 2004

[5] P Sedlmeier M Eberth D Schwarz et al ldquo+e psychologicaleffects of meditation a meta-analysisrdquo Psychological Bulletinvol 138 no 6 pp 1139ndash1171 2012

[6] M Goyal S Singh N Gould et al ldquoMeditation programs forpsychological stress and well-being a systematic review andmeta-analysisrdquo JAMA Internal Medicine vol 174 no 3pp 357ndash368 2014

[7] B Khoury B Knauper F Pagnini N Trent A Chiesa andK Carriere ldquoEmbodied mindfulnessrdquo Mindfulness vol 8no 5 pp 1ndash12 2017

[8] R Aloneftis ldquo+e role of embodiment in third wave mind-fulness based cognitive therapiesrdquo PsyPAG Quarterly vol 102pp 40ndash43 2017

[9] J M Burg T Probst T Heidenreich and J Michalak ldquoDe-velopment and psychometric evaluation of the bodymindfulnessquestionnairerdquo Mindfulness vol 8 no 3 pp 807ndash818 2017

[10] A Cebolla L Galiana D Campos et al ldquoHow does mind-fulness work Exploring a theoretical model using samples ofmeditators and non-meditatorsrdquoMindfulness pp 1ndash11 2017

[11] S Alidina Mindfulness for Dummies John Wiley amp SonsHoboken NJ USA 2014

[12] E ShoninW Van Gordon andM D Griffiths ldquoPractical tipsfor using mindfulness in general practicerdquo British Journal ofGeneral Practice vol 64 no 624 pp 368-369 2014

[13] S R Bishop M Lau S Shapiro et al ldquoMindfulness a pro-posed operational definitionrdquo Clinical Psychology Science andPractice vol 11 no 3 pp 230ndash241 2004

[14] J Kabat-Zinn ldquoSitting meditationsrdquoMindfulness vol 7 no 6pp 1441ndash1444 2016

[15] C E Jones An Investigation into the Role of Body Posture inMindfulness Practice Doctoral dissertation Canterbury ChristChurch University Canterbury UK 2016

[16] T N Hanh5eMiracle of Mindfulness Beacon Boston MAUSA 1975

[17] C P Cook-CottoneMindfulness and Yoga for Self-RegulationA Primer for Mental Health Professionals Springer PublishingCompany New York NY USA 1st edition 2015

[18] P Kaul J Passafiume R C Sargent and B F OrsquoHaraldquoMeditation acutely improves psychomotor vigilance andmay decrease sleep needrdquo Behavioral and Brain Functionsvol 6 no 1 p 1 2010

[19] A Ahani M Wahbeh M Miller et al ldquoChange in physio-logical signals during mindfulness meditationrdquo in Proceedingsof the IEEE EMBS Conference Osaka Japan July 2013

[20] J J Arch and M G Craske ldquoMechanisms of mindfulnessemotion regulation following a focused breathing inductionrdquoBehaviour Research and 5erapy vol 44 no 12 pp 1849ndash1858 2006

[21] J Vidyarthi and B E Riecke ldquoInteractively mediating ex-periences of mindfulness meditationrdquo International Journal ofHuman-Computer Studies vol 72 no 8 pp 674ndash688 2014

[22] W Y Wong and M S Wong ldquoTrunk posture monitoringwith inertial sensorsrdquo European Spine Journal vol 17 no 5pp 743ndash753 2008

[23] M Dozza L Chiari and F B Horak ldquoA portable audio-biofeedback system to improve postural controlrdquo in Pro-ceedings of the 26th Annual International Conference of the

Table 8 ANOVA results for σa (p values)

Mauchlyrsquos test of sphericitySeat 0239

Sensor location 0000Seatlowast sensor location 0000

Pillairsquos traceSeat 0675

Sensor location 0000Seatlowast sensor location 0 726

Sphericity assumedSeat 0744

Sensor location 0000Seatlowast sensor location 0725

Table 9 ANOVA results for the sway (Axz and Axy versus seat)(p value)

Axz Axy

Mauchlyrsquos test of sphericity 0000 0037Pillairsquos trace 0196 0002Sphericity assumed 0299 0006

8 Journal of Healthcare Engineering

IEEE Engineering in Medicine and Biology Society pp 4799ndash4802 San Francisco CA USA September 2004

[24] S L Whitney J L Roche G F Marchetti et al ldquoA com-parison of accelerometry and center of pressure measuresduring computerized dynamic posturography a measure ofbalancerdquo Gait and Posture vol 33 no 4 pp 594ndash599 2011

[25] W Maetzler M Mancini I Liepelt-Scarfone et al ldquoImpairedtrunk stability in individuals at high risk for Parkinsonrsquosdiseaserdquo PLoS One vol 7 no 3 article e32240 pp 1ndash6 2012

[26] J Houtan R Changbum T Ahn and L Stentz ldquoFall riskanalysis of construction workers using inertial measurementunits validating the usefulness of the postural stability metricsin constructionrdquo Safety Science vol 84 pp 161ndash170 2016

[27] S J Ozinga and J L Alberts ldquoQuantification of posturalstability in older adults using mobile technologyrdquo Experi-mental Brain Research vol 232 no 12 pp 3861ndash3972 2014

[28] S R Lord C Sherrington and H B Menz Falls in OlderPeople Risk Factors and Strategies for Prevention CambridgeUniversity Press Cambridge UK 2nd edition 2007

[29] K Chang S Chen and H Lee ldquoWireless accelerator basedbody posture stability detection and application for medita-tion practitionersrdquo in Proceedings of the International Con-ference on Biomedical and Health Informatics BHIpp 112ndash115 Hong Kong China January 2012

[30] K Chang S Chen H Lee et al ldquoA wireless accelerometer-based body posture stability detection system and its appli-cation for meditation practitionersrdquo Sensors vol 12 no 12pp 17620ndash17632 2012

[31] K M Chang Y T Chun S H Chen et al ldquo+e evaluation ofphysical stillness with wearable chest and arm accelerometerduring Chan Ding practicerdquo Sensors vol 16 p 1126 2016

[32] P M Deshmukh C M Russell L E Lucarino andS N Robinovitch ldquoEnhancing clinical measures of posturalstability with wearable sensorsrdquo in Proceedings of the 34th AnnualInternational Conference of the IEEE EMBS pp 4521ndash4524 SanDiego CA USA September 2012

[33] R P Hubble G A Naughton P A Silburn and M H ColeldquoWearable sensor use for assessing standing balance andwalking stability in people with Parkinsonrsquos disease a system-atic reviewrdquo PloS One vol 10 no 4 article e0123705 2015

[34] Lumo Web Page of Lumo Lift product 2017 httpswwwlumobodytechcomlumoliftutm_expid70259375-21Yz80XOgnSXu9lXiNTep1yA1amputm_referrerhttps3A2F2Fwwwlumobodytechcom2Fscience-of-lumo-run2F

[35] Upright Web Page of Personal Posture Trainer-UprightProduct 2017 httpwwwuprightposecom

[36] Alex Web Page of Alex Product 2017 httpalexposturecom[37] Jins-Meme Web Page of Jins Meme Glasses 2017 httpsjins-

memecomen[38] PGrossman ldquoMindfulness for psychologists paying kind attention

to the perceptiblerdquo Mindfulness vol 1 no 2 pp 87ndash97 2010[39] IMU9150 2017 httpswwwinvensensecomproductsmotion-

tracking9-axismpu-9150[40] ADXL345 2017 httpwwwanalogcomenproductsmems

accelerometersadxl345html[41] Arduino Pro Mini 2017 httpswwwarduinoccenMain

ArduinoBoardProMini[42] Bluetooth HC-05 2017 httpwwwthaieasyeleccomdownloads

EFDV390EFDV390_Datasheetpdf[43] B Najafi K Aminian A Paraschiv-Ionescu F Loew C J Bula

and P Robert ldquoAmbulatory system for human motion analysisusing a kinematic sensor monitoring of daily physical activityin the elderlyrdquo IEEE Transactions on Biomedical Engineeringvol 50 no 6 pp 711ndash723 2003

[44] K T Wyne ldquoA comprehensive review of tremorrdquo JAAPAvol 18 no 2 pp 43ndash50 2005

[45] S Gottlieb and O C J Lippold ldquo+e 4-6 Hz tremor duringsustained contraction in normal human subjectsrdquo Journal ofPhysiology vol 336 no 1 pp 499ndash509 1983

[46] M C Barroso G P Esteves T P Nunes L M G SilvaA C D Faria and P L Melo ldquoA telemedicine instrument forremote evaluation of tremor design and initial applications infatigue and patients with Parkinsonrsquos diseaserdquo BiomedicalEngineering OnLine vol 10 no 1 p 14 2011

[47] Android Developers Sensor Manager 2018 httpsdeveloperandroidcomreferenceandroidhardwareSensorManagerhtml

[48] T Deshimaru and Y Ikemi Zen et Self-Control Kairos S AEditions Retz Barcelona Spain 6th edition 1985

[49] K Sekida Zen Training Kairos S A Weatherhill Inc NewYork NY USA 6th edition 2016

[50] D Villalba What Is Zen Miraguano Ediciones MadridSpain 2005 ISBN84-7813-286-4

[51] K Noro T Naruse R Lueder N Nao-i and M KozawaldquoApplication of Zen sitting principles to microscopic surgeryseatingrdquoApplied Ergonomics vol 43 no 2 pp 308ndash319 2012

[52] K Ishihara K Dake T Kashihara and S Ishihara ldquoAn attemptto improve the sitting posture of children in the classroomsrdquo inProceedings of the International MultiConference of Engineersand Computer Scientists vol 3 Hong Kong March 2010

Journal of Healthcare Engineering 9

After recording the angles they were filtered usinga moving average filter (500 samples 10 s) to reduce noise((2) and Figure 5) which effectively removed any dynamiccomponent

y(n) 1N

1113944

iNminus1

i0x(nminus i) (2)

where N 500A global measure of the volunteerrsquos change of position

during the session was then obtained from the differencebetween the maximum and minimum values of the signalFor example the signal in Figure 5 has a variation of 1245deg

223 Statistical Analysis +e standard deviation of theacceleration was analyzed using a two-way ANOVA withrepeated measures on both factors the sensor location andthe seat (results in Section 31) Similarly to analyze theglobal change of the posture (angle differences) an ANOVAwith one factor (seat) was carried out since in this secondcase only one sensor location was used (Section 32) +estatistical analyses were performed with SPSS 120

3 Results

31 Analysis of Accelerometer Standard Deviation Twofactors have been considered in the analysis the meditationseat during the sessions (ldquoseatrdquo chair zafu or meditationbench) and the location of sensors on the body (ldquosensorlocationrdquo thumb head sternum infraclavicular region orlumbar region)

Table 1 shows the results of an ANOVA with repeatedmeasures on both factors +e details about the statisticaltests performed are given in Appendix +e sensor locationwas found to be significant (p valuelt 005) On the otherhand neither the seat nor the interaction term was signif-icant Since the sensor location was found to be significantwe proceeded with a multiple comparison analysis to dis-cover the root of the differences in terms of pairwisecomparisons of sensor locations +e p values of thecomparisons are shown in Table 2 after applying a standardBonferronirsquos correction

Table 3 shows the mean values obtained for each of thelocations of the sensors and Table 4 shows the mean valuesobtained for each of the seats (marginal values provided by

SPSS) +e marginal values are obtained by averaging overone of the factors keeping only the other one +ey can givea clue of the influence of a single factor According to Ta-bles 2 and 3 the lumbar position is the least sensitive since itsdifferences with the others are significant and themean valueis the lowest one Among the remaining positions (thumbhead sternum and infraclavicular region) the thumb showsthe highest sensitivity although the interdifferences are notsignificant except with the sternum+e marginal values forthe seat factor in Table 4 are presented for completeness andto show that the experimental values are very close

+e results of this statistical analysis remained almostthe same after preprocessing the accelerometer to lower theresolution to the worst one as explained in Section 222+e only difference is that sternum and thumb did not differsignificantly after that preprocessing

32 Posture Variation Depending on the Seat +e results ofthe ANOVA for the anterior-posterior sway (Axz) and theleft-right sway (Axy) are presented in Table 5 Table 6 showsthe values of the sway for the three seats

+e zafu seems to permit a better posture setting that canbe held throughout the session since sways were lowerHowever for Axz the effect was not significant while forAxy it was +erefore a post hoc analysis was performed forAxy +e results are presented in Table 7 +e pairwisecomparisons indicate that the zafu seat was different fromthe others

4 Discussion

Postural stability is an important factor during mindfulnessmeditation training Inertial sensors can be a good option formonitoring posture dynamics due to their low cost and easeof use One of the goals of this paper was to determine whichis the most sensitive sensor location Multiple protocols weredescribed in the literature for the placement of wearablesensors on the human body for assessing for instancestanding balance or walking stability Most of them reportedplacing a wearable sensor on either the lumbar or sacralregion of the trunk To evaluate body posture stability duringsitting meditation Chang et al [30] secured the measure-ment system between the abdomen and the chest or on thearm [31] Manufacturers of commercial products choseother body positions collarbone head lower back and soon +e lumbar region is often selected since it is close to thebody center of mass in a standing posture However theresults of the present study indicate that the lumbar or sacralregion is not the best option for measuring motion duringsitting meditation since its σa value is the lowest and sig-nificantly different from the rest It could be ruled out forfuture applications Among the remaining locations it is notpossible to separate them into different groups and only thethumb is significantly different from the sternum+ereforeconvenience or comfort can be important factors to selectthe location +e head location requires a more bulky ac-cessory +e infraclavicular region is less invasive since thesensor is attached to clothes +e sternum and lumbar could

Angle Axy

Angle Axy

Angle Axz

Angle Axz

Figure 4 Direction of rotation angles with respect to glasses Bluearrow anterior-posterior (Axz) white arrow left-right (Axy)

Journal of Healthcare Engineering 5

be more accepted by users since sensors attached to strapsare common in commercial products for tness Finally thethumb location would require a miniaturization of thesystem to improve comfortability Usability acceptance andcomfort should be evaluated among users Although thedierences are not statistically signicant the sensitivity of

the head location was higher than the sensitivity of thesternum or the infraclavicular locations is seems rea-sonable since the distance to the seat base is larger on thehead In fact the order of the sensitivity among the locationshead infraclavicular region sternum and lumbar regionfollowed the order of the distance to the seat e thumbshowed an even larger degree of movement but this could berelated to the movements of the hand that can be in-dependent of upper body displacements In fact handposture is very important according to traditional texts[48 49] and in particular thumb ngers should not drop orrise but remain in a perfect horizontal position [50]

e other goal of this paper was to analyze the inordfuenceof the seat to recommend a specic one based on dataobtained using inertial sensors In the oriental traditionmeditation is performed while seated on a cushion pref-erentially in the full-lotus or half-lotus position e staticposition is not to be changed until the end of the meditation

Table 1 ANOVA results for σa (p values)

Factor p valueSeat 0744Sensor location 0000Seatlowast sensor location 0726p values less than 005 are in bold

Table 2 Sensor location analysis pairwise comparisons withBonferronirsquos correction

umb Head Sternum Infraclavicularregion

Head 1000 mdash mdash mdashSternum 0022 0329 mdash mdashInfraclavicularregion 0787 1000 1000 mdash

Lumbar region 0000 0000 0000 0000p values less than 005 are highlighted in bold

Table 3 Sensor location marginal mean values of σa (in g)

Sensor location Mean(1) umb 00083(2) Head 00068(3) Sternum 00054(4) Infraclavicular region 00060(5) Lumbar region 00025

Table 4 Seat marginal mean values of σa (in g)

Seat Mean(1) Zafu 00059(2) Chair 00056(3) Bench 00060

Table 5 ANOVA results for the anterior-posterior sway (Axz) andthe left-right sway (Axy) versus the seat

Axz versus seat Axy versus seatp value 0196 0002p values less than 005 are highlighted in bold

Table 6 Values obtained for each seat considering anterior-pos-terior and left-right sways (Axz and Axy in degrees)

Seat Axz AxyZafu 678 (751) 583 (511)Chair 848 (862) 1001 (972)Meditation bench 1013 (1292) 918 (687)Values are given as mean (SD)

Table 7 Seat analysis of the left-right sway (Axy) pairwisecomparisons with Bonferronirsquos correction

Zafu ChairChair 0019 mdashMeditation bench 0006 1000p values less than 005 are highlighted in bold

ndash10

ndash15

ndash20

ndash25

ndash300 05 1

Samples at 50 Hz times104

Deg

rees

ndash10

ndash15

ndash20

ndash25

ndash30

Deg

rees

15 2 25 3 0 05 1Samples at 50 Hz times104

15 2 25

Figure 5 Signal before the lter (a) and signal after the lter (b)

6 Journal of Healthcare Engineering

time Western teachers and mindfulness professionals arenot so strict about this point instead of a cushion (zafu)a comfortable chair or a meditation bench can be used too Apriori one could think that the chair should give someadvantage since most of the volunteers had no meditationexperience and therefore they were not used to sit on thezafu or the meditation bench which were probably un-known for them +e only significant difference was relatedto the global change in posture throughout the session (left-right sway) +e zafu provided the best seat with significantdifferences from the chair and the meditation bench inkeeping with the oriental tradition However anterior-posterior sway and σa were not significantly different+us it might seem reasonable to advise using the zafu +isis consistent with the research in Zen sitting postures +euse of the zafu to maintain fixed postures has been applied inother fields such as in the development of a chair for mi-croscopic surgery [51] or in the improvement of the sittingposture for children in the classrooms [52]

However other aspects such as comfort should beconsidered by western and beginner meditators In thisregard some people reported pain in the anterior part of theankle when sitting on a meditation bench People with backpain or difficulty getting up from a low seated positionshould consider a comfortable chair [17]

In the context of breathing interventions this study hasserved to give a tip on the seat for meditation used in thebeginning stages of training It is supported by stabilitymeasures +e determination of sensor location is importantto reduce the number of locations to be monitored in view offuture studies in which the relation between stability andbeneficial outcomes of mindfulness should be investigatedAnyway the results of this study are preliminary and shouldbe taken with caution

One of the limitations of this work is the number ofvolunteers which corresponds to a pilot study to check thewearable sensor kit +e location of the sensors also entailssome random components due to difficulties placing them inan exact position However these differences are very lowcompared to the distance between different locations An-other concern is the fact that the accelerometers used in thisstudy include parts from different manufacturers Howeverthe measurements were taken in almost static situations andtherefore their range was irrelevant Besides we performedthe analysis described in Section 222 to lower the resolutionof the sensors to a common baseline +e main conclusionsof the statistical analyses remained the same after thispreprocessing

5 Conclusions and Future Lines of Work

Several review articles support the usefulness and effec-tiveness of mindfulness techniques in health and well-being+us many people are starting to practice mindfulness Inthe beginning stages daily practice between 5 and 15minutes is recommended in which the stable posture is a keycomponent In this work we have presented a set of inertialsensors to measure the motion and change in posture +econventionally used lumbar region is not the best body

location for sensors since it presents lower sensitivity Be-sides out of all the meditation seats the zafu has someadvantages in one out of three parameters obtained in thisstudy concerning the overall change in posture from theleft-right sway

In summary a kit for measuring stability in meditationcould consider a single sensor placed at the head if the useris willing to wear an accessory like a pair of glasses or at-tached to clothes at the infraclavicular region An additionalsensor at the thumb finger could also be used in certainforms of meditation [50] since its movements are consideredby many meditation masters as a direct indication of themeditation depth In addition beginners could start prac-ticing meditation in chairs or meditation benches althougha transition to the zafu is recommended for a medium term

+e present study could be improved in several waysFurther improvements in the orientation angles can beobtained with more complex low-pass filters or estimationwith data fusion of the accelerometer gyroscope andmagnetometer using compensatory filters +us gravitycould be determinedmore properly With all these the studycould go beyond the assumptions of a single force in thesystem (gravity) or a fixed orientation of the glasses sensorAnother hardware issue is power consumption especially iflonger sessions were to be recorded It should be measuredand reduced for instance using low-power Bluetooth (forinstance the module CC2541) +e possibility of data lossesin the Bluetooth transmission should also be considered andhandled in a future smartphone application +e measure-ments should be extended with more volunteers improvingthis pilot research In a future study the sensors will be usedto provide feedback information +e effect of this feedbackonmindfulness training programs and self-regulation will beexplored Specifically changes in the proprioception (theability to sense the position location and movement of thebody and its parts) could be evaluated with the practiceMoreover the role of movement and posture in bringingabout beneficial outcomes of mindfulness-based interven-tions should be investigated

Appendix

Detailed Description of the Statistical Test

In this section a more detailed description of the statisticaltests is given For the ANOVA analysis of σa with repeatedmeasures (factors sensor location and seat) the results areshown in Table 8 Neither the sensor location factor nor theinteraction factor fulfilled the sphericity condition so that forthem the p value of Pillairsquos trace was selected Othercommon statistics provided by SPSS like the GreenhousendashGeisser correction led to the same conclusion as Pillairsquos trace

+e ANOVA results of the analysis of the sway (Axy andAxz) with respect to the seat are shown in Table 9 No mea-surement fulfilled the sphericity test so the value of Pillairsquos tracewas selected In the case ofAxy the null hypothesis was rejectedOther common statistics of SPSS (GreenhousendashGeisser cor-rection) led to the same conclusion

Journal of Healthcare Engineering 7

Data Availability

+e data used in this article are available upon request fromthe first author

Conflicts of Interest

+e authors declare no conflicts of interest

Authorsrsquo Contributions

Inmaculada Plaza and Carlos T Medrano conceived anddesigned the experiments Victor H Rodriguez developedthe measuring kit Victor H Rodriguez and InmaculadaPlaza performed the experiments Victor H Rodriguezanalyzed the data Carlos T Medrano contributed analysistools Inmaculada Plaza Victor H Rodriguez and CarlosT Medrano wrote the paper

Acknowledgments

+e authors would like to thank all volunteers who partici-pated in the study +anks are due to ldquoCONACYT-Gobiernodel Estado de Durango Mexicordquo for the research-grant ref-erence 330795386043+is workwas supported by the ldquoFondoSocial Europeordquo the ldquoGobierno de Aragonrdquo the Ministry ofEconomy and Competitiveness (Project reference TEC2013-50049-EXP) and the University of Zaragoza (Project referenceUZ2017-TEC-02)

References

[1] J Kabat-Zinn ldquoAn outpatient program in behavioral medi-cine for chronic pain patients based on the practice ofmindfulness meditation theoretical considerations and pre-liminary resultsrdquo General Hospital Psychiatry vol 4 no 1pp 33ndash47 1982

[2] J Kabat-Zinn ldquoMindfulness-based interventions in contextpast present and futurerdquo Clinical Psychology Science andPractice vol 10 pp 144ndash156 2003

[3] R A Baer ldquoMindfulness training as clinical interventiona conceptual and empirical reviewrdquo Clinical PsychologyScience and Practice vol 10 pp 125ndash143 2003

[4] P Grossman L Niemann S Schmidt and H WalachldquoMindfulness-based stress reduction and health benefits Ameta-analysisrdquo Journal of Psychosomatic Research vol 57 no 1pp 5ndash43 2004

[5] P Sedlmeier M Eberth D Schwarz et al ldquo+e psychologicaleffects of meditation a meta-analysisrdquo Psychological Bulletinvol 138 no 6 pp 1139ndash1171 2012

[6] M Goyal S Singh N Gould et al ldquoMeditation programs forpsychological stress and well-being a systematic review andmeta-analysisrdquo JAMA Internal Medicine vol 174 no 3pp 357ndash368 2014

[7] B Khoury B Knauper F Pagnini N Trent A Chiesa andK Carriere ldquoEmbodied mindfulnessrdquo Mindfulness vol 8no 5 pp 1ndash12 2017

[8] R Aloneftis ldquo+e role of embodiment in third wave mind-fulness based cognitive therapiesrdquo PsyPAG Quarterly vol 102pp 40ndash43 2017

[9] J M Burg T Probst T Heidenreich and J Michalak ldquoDe-velopment and psychometric evaluation of the bodymindfulnessquestionnairerdquo Mindfulness vol 8 no 3 pp 807ndash818 2017

[10] A Cebolla L Galiana D Campos et al ldquoHow does mind-fulness work Exploring a theoretical model using samples ofmeditators and non-meditatorsrdquoMindfulness pp 1ndash11 2017

[11] S Alidina Mindfulness for Dummies John Wiley amp SonsHoboken NJ USA 2014

[12] E ShoninW Van Gordon andM D Griffiths ldquoPractical tipsfor using mindfulness in general practicerdquo British Journal ofGeneral Practice vol 64 no 624 pp 368-369 2014

[13] S R Bishop M Lau S Shapiro et al ldquoMindfulness a pro-posed operational definitionrdquo Clinical Psychology Science andPractice vol 11 no 3 pp 230ndash241 2004

[14] J Kabat-Zinn ldquoSitting meditationsrdquoMindfulness vol 7 no 6pp 1441ndash1444 2016

[15] C E Jones An Investigation into the Role of Body Posture inMindfulness Practice Doctoral dissertation Canterbury ChristChurch University Canterbury UK 2016

[16] T N Hanh5eMiracle of Mindfulness Beacon Boston MAUSA 1975

[17] C P Cook-CottoneMindfulness and Yoga for Self-RegulationA Primer for Mental Health Professionals Springer PublishingCompany New York NY USA 1st edition 2015

[18] P Kaul J Passafiume R C Sargent and B F OrsquoHaraldquoMeditation acutely improves psychomotor vigilance andmay decrease sleep needrdquo Behavioral and Brain Functionsvol 6 no 1 p 1 2010

[19] A Ahani M Wahbeh M Miller et al ldquoChange in physio-logical signals during mindfulness meditationrdquo in Proceedingsof the IEEE EMBS Conference Osaka Japan July 2013

[20] J J Arch and M G Craske ldquoMechanisms of mindfulnessemotion regulation following a focused breathing inductionrdquoBehaviour Research and 5erapy vol 44 no 12 pp 1849ndash1858 2006

[21] J Vidyarthi and B E Riecke ldquoInteractively mediating ex-periences of mindfulness meditationrdquo International Journal ofHuman-Computer Studies vol 72 no 8 pp 674ndash688 2014

[22] W Y Wong and M S Wong ldquoTrunk posture monitoringwith inertial sensorsrdquo European Spine Journal vol 17 no 5pp 743ndash753 2008

[23] M Dozza L Chiari and F B Horak ldquoA portable audio-biofeedback system to improve postural controlrdquo in Pro-ceedings of the 26th Annual International Conference of the

Table 8 ANOVA results for σa (p values)

Mauchlyrsquos test of sphericitySeat 0239

Sensor location 0000Seatlowast sensor location 0000

Pillairsquos traceSeat 0675

Sensor location 0000Seatlowast sensor location 0 726

Sphericity assumedSeat 0744

Sensor location 0000Seatlowast sensor location 0725

Table 9 ANOVA results for the sway (Axz and Axy versus seat)(p value)

Axz Axy

Mauchlyrsquos test of sphericity 0000 0037Pillairsquos trace 0196 0002Sphericity assumed 0299 0006

8 Journal of Healthcare Engineering

IEEE Engineering in Medicine and Biology Society pp 4799ndash4802 San Francisco CA USA September 2004

[24] S L Whitney J L Roche G F Marchetti et al ldquoA com-parison of accelerometry and center of pressure measuresduring computerized dynamic posturography a measure ofbalancerdquo Gait and Posture vol 33 no 4 pp 594ndash599 2011

[25] W Maetzler M Mancini I Liepelt-Scarfone et al ldquoImpairedtrunk stability in individuals at high risk for Parkinsonrsquosdiseaserdquo PLoS One vol 7 no 3 article e32240 pp 1ndash6 2012

[26] J Houtan R Changbum T Ahn and L Stentz ldquoFall riskanalysis of construction workers using inertial measurementunits validating the usefulness of the postural stability metricsin constructionrdquo Safety Science vol 84 pp 161ndash170 2016

[27] S J Ozinga and J L Alberts ldquoQuantification of posturalstability in older adults using mobile technologyrdquo Experi-mental Brain Research vol 232 no 12 pp 3861ndash3972 2014

[28] S R Lord C Sherrington and H B Menz Falls in OlderPeople Risk Factors and Strategies for Prevention CambridgeUniversity Press Cambridge UK 2nd edition 2007

[29] K Chang S Chen and H Lee ldquoWireless accelerator basedbody posture stability detection and application for medita-tion practitionersrdquo in Proceedings of the International Con-ference on Biomedical and Health Informatics BHIpp 112ndash115 Hong Kong China January 2012

[30] K Chang S Chen H Lee et al ldquoA wireless accelerometer-based body posture stability detection system and its appli-cation for meditation practitionersrdquo Sensors vol 12 no 12pp 17620ndash17632 2012

[31] K M Chang Y T Chun S H Chen et al ldquo+e evaluation ofphysical stillness with wearable chest and arm accelerometerduring Chan Ding practicerdquo Sensors vol 16 p 1126 2016

[32] P M Deshmukh C M Russell L E Lucarino andS N Robinovitch ldquoEnhancing clinical measures of posturalstability with wearable sensorsrdquo in Proceedings of the 34th AnnualInternational Conference of the IEEE EMBS pp 4521ndash4524 SanDiego CA USA September 2012

[33] R P Hubble G A Naughton P A Silburn and M H ColeldquoWearable sensor use for assessing standing balance andwalking stability in people with Parkinsonrsquos disease a system-atic reviewrdquo PloS One vol 10 no 4 article e0123705 2015

[34] Lumo Web Page of Lumo Lift product 2017 httpswwwlumobodytechcomlumoliftutm_expid70259375-21Yz80XOgnSXu9lXiNTep1yA1amputm_referrerhttps3A2F2Fwwwlumobodytechcom2Fscience-of-lumo-run2F

[35] Upright Web Page of Personal Posture Trainer-UprightProduct 2017 httpwwwuprightposecom

[36] Alex Web Page of Alex Product 2017 httpalexposturecom[37] Jins-Meme Web Page of Jins Meme Glasses 2017 httpsjins-

memecomen[38] PGrossman ldquoMindfulness for psychologists paying kind attention

to the perceptiblerdquo Mindfulness vol 1 no 2 pp 87ndash97 2010[39] IMU9150 2017 httpswwwinvensensecomproductsmotion-

tracking9-axismpu-9150[40] ADXL345 2017 httpwwwanalogcomenproductsmems

accelerometersadxl345html[41] Arduino Pro Mini 2017 httpswwwarduinoccenMain

ArduinoBoardProMini[42] Bluetooth HC-05 2017 httpwwwthaieasyeleccomdownloads

EFDV390EFDV390_Datasheetpdf[43] B Najafi K Aminian A Paraschiv-Ionescu F Loew C J Bula

and P Robert ldquoAmbulatory system for human motion analysisusing a kinematic sensor monitoring of daily physical activityin the elderlyrdquo IEEE Transactions on Biomedical Engineeringvol 50 no 6 pp 711ndash723 2003

[44] K T Wyne ldquoA comprehensive review of tremorrdquo JAAPAvol 18 no 2 pp 43ndash50 2005

[45] S Gottlieb and O C J Lippold ldquo+e 4-6 Hz tremor duringsustained contraction in normal human subjectsrdquo Journal ofPhysiology vol 336 no 1 pp 499ndash509 1983

[46] M C Barroso G P Esteves T P Nunes L M G SilvaA C D Faria and P L Melo ldquoA telemedicine instrument forremote evaluation of tremor design and initial applications infatigue and patients with Parkinsonrsquos diseaserdquo BiomedicalEngineering OnLine vol 10 no 1 p 14 2011

[47] Android Developers Sensor Manager 2018 httpsdeveloperandroidcomreferenceandroidhardwareSensorManagerhtml

[48] T Deshimaru and Y Ikemi Zen et Self-Control Kairos S AEditions Retz Barcelona Spain 6th edition 1985

[49] K Sekida Zen Training Kairos S A Weatherhill Inc NewYork NY USA 6th edition 2016

[50] D Villalba What Is Zen Miraguano Ediciones MadridSpain 2005 ISBN84-7813-286-4

[51] K Noro T Naruse R Lueder N Nao-i and M KozawaldquoApplication of Zen sitting principles to microscopic surgeryseatingrdquoApplied Ergonomics vol 43 no 2 pp 308ndash319 2012

[52] K Ishihara K Dake T Kashihara and S Ishihara ldquoAn attemptto improve the sitting posture of children in the classroomsrdquo inProceedings of the International MultiConference of Engineersand Computer Scientists vol 3 Hong Kong March 2010

Journal of Healthcare Engineering 9

be more accepted by users since sensors attached to strapsare common in commercial products for tness Finally thethumb location would require a miniaturization of thesystem to improve comfortability Usability acceptance andcomfort should be evaluated among users Although thedierences are not statistically signicant the sensitivity of

the head location was higher than the sensitivity of thesternum or the infraclavicular locations is seems rea-sonable since the distance to the seat base is larger on thehead In fact the order of the sensitivity among the locationshead infraclavicular region sternum and lumbar regionfollowed the order of the distance to the seat e thumbshowed an even larger degree of movement but this could berelated to the movements of the hand that can be in-dependent of upper body displacements In fact handposture is very important according to traditional texts[48 49] and in particular thumb ngers should not drop orrise but remain in a perfect horizontal position [50]

e other goal of this paper was to analyze the inordfuenceof the seat to recommend a specic one based on dataobtained using inertial sensors In the oriental traditionmeditation is performed while seated on a cushion pref-erentially in the full-lotus or half-lotus position e staticposition is not to be changed until the end of the meditation

Table 1 ANOVA results for σa (p values)

Factor p valueSeat 0744Sensor location 0000Seatlowast sensor location 0726p values less than 005 are in bold

Table 2 Sensor location analysis pairwise comparisons withBonferronirsquos correction

umb Head Sternum Infraclavicularregion

Head 1000 mdash mdash mdashSternum 0022 0329 mdash mdashInfraclavicularregion 0787 1000 1000 mdash

Lumbar region 0000 0000 0000 0000p values less than 005 are highlighted in bold

Table 3 Sensor location marginal mean values of σa (in g)

Sensor location Mean(1) umb 00083(2) Head 00068(3) Sternum 00054(4) Infraclavicular region 00060(5) Lumbar region 00025

Table 4 Seat marginal mean values of σa (in g)

Seat Mean(1) Zafu 00059(2) Chair 00056(3) Bench 00060

Table 5 ANOVA results for the anterior-posterior sway (Axz) andthe left-right sway (Axy) versus the seat

Axz versus seat Axy versus seatp value 0196 0002p values less than 005 are highlighted in bold

Table 6 Values obtained for each seat considering anterior-pos-terior and left-right sways (Axz and Axy in degrees)

Seat Axz AxyZafu 678 (751) 583 (511)Chair 848 (862) 1001 (972)Meditation bench 1013 (1292) 918 (687)Values are given as mean (SD)

Table 7 Seat analysis of the left-right sway (Axy) pairwisecomparisons with Bonferronirsquos correction

Zafu ChairChair 0019 mdashMeditation bench 0006 1000p values less than 005 are highlighted in bold

ndash10

ndash15

ndash20

ndash25

ndash300 05 1

Samples at 50 Hz times104

Deg

rees

ndash10

ndash15

ndash20

ndash25

ndash30

Deg

rees

15 2 25 3 0 05 1Samples at 50 Hz times104

15 2 25

Figure 5 Signal before the lter (a) and signal after the lter (b)

6 Journal of Healthcare Engineering

time Western teachers and mindfulness professionals arenot so strict about this point instead of a cushion (zafu)a comfortable chair or a meditation bench can be used too Apriori one could think that the chair should give someadvantage since most of the volunteers had no meditationexperience and therefore they were not used to sit on thezafu or the meditation bench which were probably un-known for them +e only significant difference was relatedto the global change in posture throughout the session (left-right sway) +e zafu provided the best seat with significantdifferences from the chair and the meditation bench inkeeping with the oriental tradition However anterior-posterior sway and σa were not significantly different+us it might seem reasonable to advise using the zafu +isis consistent with the research in Zen sitting postures +euse of the zafu to maintain fixed postures has been applied inother fields such as in the development of a chair for mi-croscopic surgery [51] or in the improvement of the sittingposture for children in the classrooms [52]

However other aspects such as comfort should beconsidered by western and beginner meditators In thisregard some people reported pain in the anterior part of theankle when sitting on a meditation bench People with backpain or difficulty getting up from a low seated positionshould consider a comfortable chair [17]

In the context of breathing interventions this study hasserved to give a tip on the seat for meditation used in thebeginning stages of training It is supported by stabilitymeasures +e determination of sensor location is importantto reduce the number of locations to be monitored in view offuture studies in which the relation between stability andbeneficial outcomes of mindfulness should be investigatedAnyway the results of this study are preliminary and shouldbe taken with caution

One of the limitations of this work is the number ofvolunteers which corresponds to a pilot study to check thewearable sensor kit +e location of the sensors also entailssome random components due to difficulties placing them inan exact position However these differences are very lowcompared to the distance between different locations An-other concern is the fact that the accelerometers used in thisstudy include parts from different manufacturers Howeverthe measurements were taken in almost static situations andtherefore their range was irrelevant Besides we performedthe analysis described in Section 222 to lower the resolutionof the sensors to a common baseline +e main conclusionsof the statistical analyses remained the same after thispreprocessing

5 Conclusions and Future Lines of Work

Several review articles support the usefulness and effec-tiveness of mindfulness techniques in health and well-being+us many people are starting to practice mindfulness Inthe beginning stages daily practice between 5 and 15minutes is recommended in which the stable posture is a keycomponent In this work we have presented a set of inertialsensors to measure the motion and change in posture +econventionally used lumbar region is not the best body

location for sensors since it presents lower sensitivity Be-sides out of all the meditation seats the zafu has someadvantages in one out of three parameters obtained in thisstudy concerning the overall change in posture from theleft-right sway

In summary a kit for measuring stability in meditationcould consider a single sensor placed at the head if the useris willing to wear an accessory like a pair of glasses or at-tached to clothes at the infraclavicular region An additionalsensor at the thumb finger could also be used in certainforms of meditation [50] since its movements are consideredby many meditation masters as a direct indication of themeditation depth In addition beginners could start prac-ticing meditation in chairs or meditation benches althougha transition to the zafu is recommended for a medium term

+e present study could be improved in several waysFurther improvements in the orientation angles can beobtained with more complex low-pass filters or estimationwith data fusion of the accelerometer gyroscope andmagnetometer using compensatory filters +us gravitycould be determinedmore properly With all these the studycould go beyond the assumptions of a single force in thesystem (gravity) or a fixed orientation of the glasses sensorAnother hardware issue is power consumption especially iflonger sessions were to be recorded It should be measuredand reduced for instance using low-power Bluetooth (forinstance the module CC2541) +e possibility of data lossesin the Bluetooth transmission should also be considered andhandled in a future smartphone application +e measure-ments should be extended with more volunteers improvingthis pilot research In a future study the sensors will be usedto provide feedback information +e effect of this feedbackonmindfulness training programs and self-regulation will beexplored Specifically changes in the proprioception (theability to sense the position location and movement of thebody and its parts) could be evaluated with the practiceMoreover the role of movement and posture in bringingabout beneficial outcomes of mindfulness-based interven-tions should be investigated

Appendix

Detailed Description of the Statistical Test

In this section a more detailed description of the statisticaltests is given For the ANOVA analysis of σa with repeatedmeasures (factors sensor location and seat) the results areshown in Table 8 Neither the sensor location factor nor theinteraction factor fulfilled the sphericity condition so that forthem the p value of Pillairsquos trace was selected Othercommon statistics provided by SPSS like the GreenhousendashGeisser correction led to the same conclusion as Pillairsquos trace

+e ANOVA results of the analysis of the sway (Axy andAxz) with respect to the seat are shown in Table 9 No mea-surement fulfilled the sphericity test so the value of Pillairsquos tracewas selected In the case ofAxy the null hypothesis was rejectedOther common statistics of SPSS (GreenhousendashGeisser cor-rection) led to the same conclusion

Journal of Healthcare Engineering 7

Data Availability

+e data used in this article are available upon request fromthe first author

Conflicts of Interest

+e authors declare no conflicts of interest

Authorsrsquo Contributions

Inmaculada Plaza and Carlos T Medrano conceived anddesigned the experiments Victor H Rodriguez developedthe measuring kit Victor H Rodriguez and InmaculadaPlaza performed the experiments Victor H Rodriguezanalyzed the data Carlos T Medrano contributed analysistools Inmaculada Plaza Victor H Rodriguez and CarlosT Medrano wrote the paper

Acknowledgments

+e authors would like to thank all volunteers who partici-pated in the study +anks are due to ldquoCONACYT-Gobiernodel Estado de Durango Mexicordquo for the research-grant ref-erence 330795386043+is workwas supported by the ldquoFondoSocial Europeordquo the ldquoGobierno de Aragonrdquo the Ministry ofEconomy and Competitiveness (Project reference TEC2013-50049-EXP) and the University of Zaragoza (Project referenceUZ2017-TEC-02)

References

[1] J Kabat-Zinn ldquoAn outpatient program in behavioral medi-cine for chronic pain patients based on the practice ofmindfulness meditation theoretical considerations and pre-liminary resultsrdquo General Hospital Psychiatry vol 4 no 1pp 33ndash47 1982

[2] J Kabat-Zinn ldquoMindfulness-based interventions in contextpast present and futurerdquo Clinical Psychology Science andPractice vol 10 pp 144ndash156 2003

[3] R A Baer ldquoMindfulness training as clinical interventiona conceptual and empirical reviewrdquo Clinical PsychologyScience and Practice vol 10 pp 125ndash143 2003

[4] P Grossman L Niemann S Schmidt and H WalachldquoMindfulness-based stress reduction and health benefits Ameta-analysisrdquo Journal of Psychosomatic Research vol 57 no 1pp 5ndash43 2004

[5] P Sedlmeier M Eberth D Schwarz et al ldquo+e psychologicaleffects of meditation a meta-analysisrdquo Psychological Bulletinvol 138 no 6 pp 1139ndash1171 2012

[6] M Goyal S Singh N Gould et al ldquoMeditation programs forpsychological stress and well-being a systematic review andmeta-analysisrdquo JAMA Internal Medicine vol 174 no 3pp 357ndash368 2014

[7] B Khoury B Knauper F Pagnini N Trent A Chiesa andK Carriere ldquoEmbodied mindfulnessrdquo Mindfulness vol 8no 5 pp 1ndash12 2017

[8] R Aloneftis ldquo+e role of embodiment in third wave mind-fulness based cognitive therapiesrdquo PsyPAG Quarterly vol 102pp 40ndash43 2017

[9] J M Burg T Probst T Heidenreich and J Michalak ldquoDe-velopment and psychometric evaluation of the bodymindfulnessquestionnairerdquo Mindfulness vol 8 no 3 pp 807ndash818 2017

[10] A Cebolla L Galiana D Campos et al ldquoHow does mind-fulness work Exploring a theoretical model using samples ofmeditators and non-meditatorsrdquoMindfulness pp 1ndash11 2017

[11] S Alidina Mindfulness for Dummies John Wiley amp SonsHoboken NJ USA 2014

[12] E ShoninW Van Gordon andM D Griffiths ldquoPractical tipsfor using mindfulness in general practicerdquo British Journal ofGeneral Practice vol 64 no 624 pp 368-369 2014

[13] S R Bishop M Lau S Shapiro et al ldquoMindfulness a pro-posed operational definitionrdquo Clinical Psychology Science andPractice vol 11 no 3 pp 230ndash241 2004

[14] J Kabat-Zinn ldquoSitting meditationsrdquoMindfulness vol 7 no 6pp 1441ndash1444 2016

[15] C E Jones An Investigation into the Role of Body Posture inMindfulness Practice Doctoral dissertation Canterbury ChristChurch University Canterbury UK 2016

[16] T N Hanh5eMiracle of Mindfulness Beacon Boston MAUSA 1975

[17] C P Cook-CottoneMindfulness and Yoga for Self-RegulationA Primer for Mental Health Professionals Springer PublishingCompany New York NY USA 1st edition 2015

[18] P Kaul J Passafiume R C Sargent and B F OrsquoHaraldquoMeditation acutely improves psychomotor vigilance andmay decrease sleep needrdquo Behavioral and Brain Functionsvol 6 no 1 p 1 2010

[19] A Ahani M Wahbeh M Miller et al ldquoChange in physio-logical signals during mindfulness meditationrdquo in Proceedingsof the IEEE EMBS Conference Osaka Japan July 2013

[20] J J Arch and M G Craske ldquoMechanisms of mindfulnessemotion regulation following a focused breathing inductionrdquoBehaviour Research and 5erapy vol 44 no 12 pp 1849ndash1858 2006

[21] J Vidyarthi and B E Riecke ldquoInteractively mediating ex-periences of mindfulness meditationrdquo International Journal ofHuman-Computer Studies vol 72 no 8 pp 674ndash688 2014

[22] W Y Wong and M S Wong ldquoTrunk posture monitoringwith inertial sensorsrdquo European Spine Journal vol 17 no 5pp 743ndash753 2008

[23] M Dozza L Chiari and F B Horak ldquoA portable audio-biofeedback system to improve postural controlrdquo in Pro-ceedings of the 26th Annual International Conference of the

Table 8 ANOVA results for σa (p values)

Mauchlyrsquos test of sphericitySeat 0239

Sensor location 0000Seatlowast sensor location 0000

Pillairsquos traceSeat 0675

Sensor location 0000Seatlowast sensor location 0 726

Sphericity assumedSeat 0744

Sensor location 0000Seatlowast sensor location 0725

Table 9 ANOVA results for the sway (Axz and Axy versus seat)(p value)

Axz Axy

Mauchlyrsquos test of sphericity 0000 0037Pillairsquos trace 0196 0002Sphericity assumed 0299 0006

8 Journal of Healthcare Engineering

IEEE Engineering in Medicine and Biology Society pp 4799ndash4802 San Francisco CA USA September 2004

[24] S L Whitney J L Roche G F Marchetti et al ldquoA com-parison of accelerometry and center of pressure measuresduring computerized dynamic posturography a measure ofbalancerdquo Gait and Posture vol 33 no 4 pp 594ndash599 2011

[25] W Maetzler M Mancini I Liepelt-Scarfone et al ldquoImpairedtrunk stability in individuals at high risk for Parkinsonrsquosdiseaserdquo PLoS One vol 7 no 3 article e32240 pp 1ndash6 2012

[26] J Houtan R Changbum T Ahn and L Stentz ldquoFall riskanalysis of construction workers using inertial measurementunits validating the usefulness of the postural stability metricsin constructionrdquo Safety Science vol 84 pp 161ndash170 2016

[27] S J Ozinga and J L Alberts ldquoQuantification of posturalstability in older adults using mobile technologyrdquo Experi-mental Brain Research vol 232 no 12 pp 3861ndash3972 2014

[28] S R Lord C Sherrington and H B Menz Falls in OlderPeople Risk Factors and Strategies for Prevention CambridgeUniversity Press Cambridge UK 2nd edition 2007

[29] K Chang S Chen and H Lee ldquoWireless accelerator basedbody posture stability detection and application for medita-tion practitionersrdquo in Proceedings of the International Con-ference on Biomedical and Health Informatics BHIpp 112ndash115 Hong Kong China January 2012

[30] K Chang S Chen H Lee et al ldquoA wireless accelerometer-based body posture stability detection system and its appli-cation for meditation practitionersrdquo Sensors vol 12 no 12pp 17620ndash17632 2012

[31] K M Chang Y T Chun S H Chen et al ldquo+e evaluation ofphysical stillness with wearable chest and arm accelerometerduring Chan Ding practicerdquo Sensors vol 16 p 1126 2016

[32] P M Deshmukh C M Russell L E Lucarino andS N Robinovitch ldquoEnhancing clinical measures of posturalstability with wearable sensorsrdquo in Proceedings of the 34th AnnualInternational Conference of the IEEE EMBS pp 4521ndash4524 SanDiego CA USA September 2012

[33] R P Hubble G A Naughton P A Silburn and M H ColeldquoWearable sensor use for assessing standing balance andwalking stability in people with Parkinsonrsquos disease a system-atic reviewrdquo PloS One vol 10 no 4 article e0123705 2015

[34] Lumo Web Page of Lumo Lift product 2017 httpswwwlumobodytechcomlumoliftutm_expid70259375-21Yz80XOgnSXu9lXiNTep1yA1amputm_referrerhttps3A2F2Fwwwlumobodytechcom2Fscience-of-lumo-run2F

[35] Upright Web Page of Personal Posture Trainer-UprightProduct 2017 httpwwwuprightposecom

[36] Alex Web Page of Alex Product 2017 httpalexposturecom[37] Jins-Meme Web Page of Jins Meme Glasses 2017 httpsjins-

memecomen[38] PGrossman ldquoMindfulness for psychologists paying kind attention

to the perceptiblerdquo Mindfulness vol 1 no 2 pp 87ndash97 2010[39] IMU9150 2017 httpswwwinvensensecomproductsmotion-

tracking9-axismpu-9150[40] ADXL345 2017 httpwwwanalogcomenproductsmems

accelerometersadxl345html[41] Arduino Pro Mini 2017 httpswwwarduinoccenMain

ArduinoBoardProMini[42] Bluetooth HC-05 2017 httpwwwthaieasyeleccomdownloads

EFDV390EFDV390_Datasheetpdf[43] B Najafi K Aminian A Paraschiv-Ionescu F Loew C J Bula

and P Robert ldquoAmbulatory system for human motion analysisusing a kinematic sensor monitoring of daily physical activityin the elderlyrdquo IEEE Transactions on Biomedical Engineeringvol 50 no 6 pp 711ndash723 2003

[44] K T Wyne ldquoA comprehensive review of tremorrdquo JAAPAvol 18 no 2 pp 43ndash50 2005

[45] S Gottlieb and O C J Lippold ldquo+e 4-6 Hz tremor duringsustained contraction in normal human subjectsrdquo Journal ofPhysiology vol 336 no 1 pp 499ndash509 1983

[46] M C Barroso G P Esteves T P Nunes L M G SilvaA C D Faria and P L Melo ldquoA telemedicine instrument forremote evaluation of tremor design and initial applications infatigue and patients with Parkinsonrsquos diseaserdquo BiomedicalEngineering OnLine vol 10 no 1 p 14 2011

[47] Android Developers Sensor Manager 2018 httpsdeveloperandroidcomreferenceandroidhardwareSensorManagerhtml

[48] T Deshimaru and Y Ikemi Zen et Self-Control Kairos S AEditions Retz Barcelona Spain 6th edition 1985

[49] K Sekida Zen Training Kairos S A Weatherhill Inc NewYork NY USA 6th edition 2016

[50] D Villalba What Is Zen Miraguano Ediciones MadridSpain 2005 ISBN84-7813-286-4

[51] K Noro T Naruse R Lueder N Nao-i and M KozawaldquoApplication of Zen sitting principles to microscopic surgeryseatingrdquoApplied Ergonomics vol 43 no 2 pp 308ndash319 2012

[52] K Ishihara K Dake T Kashihara and S Ishihara ldquoAn attemptto improve the sitting posture of children in the classroomsrdquo inProceedings of the International MultiConference of Engineersand Computer Scientists vol 3 Hong Kong March 2010

Journal of Healthcare Engineering 9

time Western teachers and mindfulness professionals arenot so strict about this point instead of a cushion (zafu)a comfortable chair or a meditation bench can be used too Apriori one could think that the chair should give someadvantage since most of the volunteers had no meditationexperience and therefore they were not used to sit on thezafu or the meditation bench which were probably un-known for them +e only significant difference was relatedto the global change in posture throughout the session (left-right sway) +e zafu provided the best seat with significantdifferences from the chair and the meditation bench inkeeping with the oriental tradition However anterior-posterior sway and σa were not significantly different+us it might seem reasonable to advise using the zafu +isis consistent with the research in Zen sitting postures +euse of the zafu to maintain fixed postures has been applied inother fields such as in the development of a chair for mi-croscopic surgery [51] or in the improvement of the sittingposture for children in the classrooms [52]

However other aspects such as comfort should beconsidered by western and beginner meditators In thisregard some people reported pain in the anterior part of theankle when sitting on a meditation bench People with backpain or difficulty getting up from a low seated positionshould consider a comfortable chair [17]

In the context of breathing interventions this study hasserved to give a tip on the seat for meditation used in thebeginning stages of training It is supported by stabilitymeasures +e determination of sensor location is importantto reduce the number of locations to be monitored in view offuture studies in which the relation between stability andbeneficial outcomes of mindfulness should be investigatedAnyway the results of this study are preliminary and shouldbe taken with caution

One of the limitations of this work is the number ofvolunteers which corresponds to a pilot study to check thewearable sensor kit +e location of the sensors also entailssome random components due to difficulties placing them inan exact position However these differences are very lowcompared to the distance between different locations An-other concern is the fact that the accelerometers used in thisstudy include parts from different manufacturers Howeverthe measurements were taken in almost static situations andtherefore their range was irrelevant Besides we performedthe analysis described in Section 222 to lower the resolutionof the sensors to a common baseline +e main conclusionsof the statistical analyses remained the same after thispreprocessing

5 Conclusions and Future Lines of Work

Several review articles support the usefulness and effec-tiveness of mindfulness techniques in health and well-being+us many people are starting to practice mindfulness Inthe beginning stages daily practice between 5 and 15minutes is recommended in which the stable posture is a keycomponent In this work we have presented a set of inertialsensors to measure the motion and change in posture +econventionally used lumbar region is not the best body

location for sensors since it presents lower sensitivity Be-sides out of all the meditation seats the zafu has someadvantages in one out of three parameters obtained in thisstudy concerning the overall change in posture from theleft-right sway

In summary a kit for measuring stability in meditationcould consider a single sensor placed at the head if the useris willing to wear an accessory like a pair of glasses or at-tached to clothes at the infraclavicular region An additionalsensor at the thumb finger could also be used in certainforms of meditation [50] since its movements are consideredby many meditation masters as a direct indication of themeditation depth In addition beginners could start prac-ticing meditation in chairs or meditation benches althougha transition to the zafu is recommended for a medium term

+e present study could be improved in several waysFurther improvements in the orientation angles can beobtained with more complex low-pass filters or estimationwith data fusion of the accelerometer gyroscope andmagnetometer using compensatory filters +us gravitycould be determinedmore properly With all these the studycould go beyond the assumptions of a single force in thesystem (gravity) or a fixed orientation of the glasses sensorAnother hardware issue is power consumption especially iflonger sessions were to be recorded It should be measuredand reduced for instance using low-power Bluetooth (forinstance the module CC2541) +e possibility of data lossesin the Bluetooth transmission should also be considered andhandled in a future smartphone application +e measure-ments should be extended with more volunteers improvingthis pilot research In a future study the sensors will be usedto provide feedback information +e effect of this feedbackonmindfulness training programs and self-regulation will beexplored Specifically changes in the proprioception (theability to sense the position location and movement of thebody and its parts) could be evaluated with the practiceMoreover the role of movement and posture in bringingabout beneficial outcomes of mindfulness-based interven-tions should be investigated

Appendix

Detailed Description of the Statistical Test

In this section a more detailed description of the statisticaltests is given For the ANOVA analysis of σa with repeatedmeasures (factors sensor location and seat) the results areshown in Table 8 Neither the sensor location factor nor theinteraction factor fulfilled the sphericity condition so that forthem the p value of Pillairsquos trace was selected Othercommon statistics provided by SPSS like the GreenhousendashGeisser correction led to the same conclusion as Pillairsquos trace

+e ANOVA results of the analysis of the sway (Axy andAxz) with respect to the seat are shown in Table 9 No mea-surement fulfilled the sphericity test so the value of Pillairsquos tracewas selected In the case ofAxy the null hypothesis was rejectedOther common statistics of SPSS (GreenhousendashGeisser cor-rection) led to the same conclusion

Journal of Healthcare Engineering 7

Data Availability

+e data used in this article are available upon request fromthe first author

Conflicts of Interest

+e authors declare no conflicts of interest

Authorsrsquo Contributions

Inmaculada Plaza and Carlos T Medrano conceived anddesigned the experiments Victor H Rodriguez developedthe measuring kit Victor H Rodriguez and InmaculadaPlaza performed the experiments Victor H Rodriguezanalyzed the data Carlos T Medrano contributed analysistools Inmaculada Plaza Victor H Rodriguez and CarlosT Medrano wrote the paper

Acknowledgments

+e authors would like to thank all volunteers who partici-pated in the study +anks are due to ldquoCONACYT-Gobiernodel Estado de Durango Mexicordquo for the research-grant ref-erence 330795386043+is workwas supported by the ldquoFondoSocial Europeordquo the ldquoGobierno de Aragonrdquo the Ministry ofEconomy and Competitiveness (Project reference TEC2013-50049-EXP) and the University of Zaragoza (Project referenceUZ2017-TEC-02)

References

[1] J Kabat-Zinn ldquoAn outpatient program in behavioral medi-cine for chronic pain patients based on the practice ofmindfulness meditation theoretical considerations and pre-liminary resultsrdquo General Hospital Psychiatry vol 4 no 1pp 33ndash47 1982

[2] J Kabat-Zinn ldquoMindfulness-based interventions in contextpast present and futurerdquo Clinical Psychology Science andPractice vol 10 pp 144ndash156 2003

[3] R A Baer ldquoMindfulness training as clinical interventiona conceptual and empirical reviewrdquo Clinical PsychologyScience and Practice vol 10 pp 125ndash143 2003

[4] P Grossman L Niemann S Schmidt and H WalachldquoMindfulness-based stress reduction and health benefits Ameta-analysisrdquo Journal of Psychosomatic Research vol 57 no 1pp 5ndash43 2004

[5] P Sedlmeier M Eberth D Schwarz et al ldquo+e psychologicaleffects of meditation a meta-analysisrdquo Psychological Bulletinvol 138 no 6 pp 1139ndash1171 2012

[6] M Goyal S Singh N Gould et al ldquoMeditation programs forpsychological stress and well-being a systematic review andmeta-analysisrdquo JAMA Internal Medicine vol 174 no 3pp 357ndash368 2014

[7] B Khoury B Knauper F Pagnini N Trent A Chiesa andK Carriere ldquoEmbodied mindfulnessrdquo Mindfulness vol 8no 5 pp 1ndash12 2017

[8] R Aloneftis ldquo+e role of embodiment in third wave mind-fulness based cognitive therapiesrdquo PsyPAG Quarterly vol 102pp 40ndash43 2017

[9] J M Burg T Probst T Heidenreich and J Michalak ldquoDe-velopment and psychometric evaluation of the bodymindfulnessquestionnairerdquo Mindfulness vol 8 no 3 pp 807ndash818 2017

[10] A Cebolla L Galiana D Campos et al ldquoHow does mind-fulness work Exploring a theoretical model using samples ofmeditators and non-meditatorsrdquoMindfulness pp 1ndash11 2017

[11] S Alidina Mindfulness for Dummies John Wiley amp SonsHoboken NJ USA 2014

[12] E ShoninW Van Gordon andM D Griffiths ldquoPractical tipsfor using mindfulness in general practicerdquo British Journal ofGeneral Practice vol 64 no 624 pp 368-369 2014

[13] S R Bishop M Lau S Shapiro et al ldquoMindfulness a pro-posed operational definitionrdquo Clinical Psychology Science andPractice vol 11 no 3 pp 230ndash241 2004

[14] J Kabat-Zinn ldquoSitting meditationsrdquoMindfulness vol 7 no 6pp 1441ndash1444 2016

[15] C E Jones An Investigation into the Role of Body Posture inMindfulness Practice Doctoral dissertation Canterbury ChristChurch University Canterbury UK 2016

[16] T N Hanh5eMiracle of Mindfulness Beacon Boston MAUSA 1975

[17] C P Cook-CottoneMindfulness and Yoga for Self-RegulationA Primer for Mental Health Professionals Springer PublishingCompany New York NY USA 1st edition 2015

[18] P Kaul J Passafiume R C Sargent and B F OrsquoHaraldquoMeditation acutely improves psychomotor vigilance andmay decrease sleep needrdquo Behavioral and Brain Functionsvol 6 no 1 p 1 2010

[19] A Ahani M Wahbeh M Miller et al ldquoChange in physio-logical signals during mindfulness meditationrdquo in Proceedingsof the IEEE EMBS Conference Osaka Japan July 2013

[20] J J Arch and M G Craske ldquoMechanisms of mindfulnessemotion regulation following a focused breathing inductionrdquoBehaviour Research and 5erapy vol 44 no 12 pp 1849ndash1858 2006

[21] J Vidyarthi and B E Riecke ldquoInteractively mediating ex-periences of mindfulness meditationrdquo International Journal ofHuman-Computer Studies vol 72 no 8 pp 674ndash688 2014

[22] W Y Wong and M S Wong ldquoTrunk posture monitoringwith inertial sensorsrdquo European Spine Journal vol 17 no 5pp 743ndash753 2008

[23] M Dozza L Chiari and F B Horak ldquoA portable audio-biofeedback system to improve postural controlrdquo in Pro-ceedings of the 26th Annual International Conference of the

Table 8 ANOVA results for σa (p values)

Mauchlyrsquos test of sphericitySeat 0239

Sensor location 0000Seatlowast sensor location 0000

Pillairsquos traceSeat 0675

Sensor location 0000Seatlowast sensor location 0 726

Sphericity assumedSeat 0744

Sensor location 0000Seatlowast sensor location 0725

Table 9 ANOVA results for the sway (Axz and Axy versus seat)(p value)

Axz Axy

Mauchlyrsquos test of sphericity 0000 0037Pillairsquos trace 0196 0002Sphericity assumed 0299 0006

8 Journal of Healthcare Engineering

IEEE Engineering in Medicine and Biology Society pp 4799ndash4802 San Francisco CA USA September 2004

[24] S L Whitney J L Roche G F Marchetti et al ldquoA com-parison of accelerometry and center of pressure measuresduring computerized dynamic posturography a measure ofbalancerdquo Gait and Posture vol 33 no 4 pp 594ndash599 2011

[25] W Maetzler M Mancini I Liepelt-Scarfone et al ldquoImpairedtrunk stability in individuals at high risk for Parkinsonrsquosdiseaserdquo PLoS One vol 7 no 3 article e32240 pp 1ndash6 2012

[26] J Houtan R Changbum T Ahn and L Stentz ldquoFall riskanalysis of construction workers using inertial measurementunits validating the usefulness of the postural stability metricsin constructionrdquo Safety Science vol 84 pp 161ndash170 2016

[27] S J Ozinga and J L Alberts ldquoQuantification of posturalstability in older adults using mobile technologyrdquo Experi-mental Brain Research vol 232 no 12 pp 3861ndash3972 2014

[28] S R Lord C Sherrington and H B Menz Falls in OlderPeople Risk Factors and Strategies for Prevention CambridgeUniversity Press Cambridge UK 2nd edition 2007

[29] K Chang S Chen and H Lee ldquoWireless accelerator basedbody posture stability detection and application for medita-tion practitionersrdquo in Proceedings of the International Con-ference on Biomedical and Health Informatics BHIpp 112ndash115 Hong Kong China January 2012

[30] K Chang S Chen H Lee et al ldquoA wireless accelerometer-based body posture stability detection system and its appli-cation for meditation practitionersrdquo Sensors vol 12 no 12pp 17620ndash17632 2012

[31] K M Chang Y T Chun S H Chen et al ldquo+e evaluation ofphysical stillness with wearable chest and arm accelerometerduring Chan Ding practicerdquo Sensors vol 16 p 1126 2016

[32] P M Deshmukh C M Russell L E Lucarino andS N Robinovitch ldquoEnhancing clinical measures of posturalstability with wearable sensorsrdquo in Proceedings of the 34th AnnualInternational Conference of the IEEE EMBS pp 4521ndash4524 SanDiego CA USA September 2012

[33] R P Hubble G A Naughton P A Silburn and M H ColeldquoWearable sensor use for assessing standing balance andwalking stability in people with Parkinsonrsquos disease a system-atic reviewrdquo PloS One vol 10 no 4 article e0123705 2015

[34] Lumo Web Page of Lumo Lift product 2017 httpswwwlumobodytechcomlumoliftutm_expid70259375-21Yz80XOgnSXu9lXiNTep1yA1amputm_referrerhttps3A2F2Fwwwlumobodytechcom2Fscience-of-lumo-run2F

[35] Upright Web Page of Personal Posture Trainer-UprightProduct 2017 httpwwwuprightposecom

[36] Alex Web Page of Alex Product 2017 httpalexposturecom[37] Jins-Meme Web Page of Jins Meme Glasses 2017 httpsjins-

memecomen[38] PGrossman ldquoMindfulness for psychologists paying kind attention

to the perceptiblerdquo Mindfulness vol 1 no 2 pp 87ndash97 2010[39] IMU9150 2017 httpswwwinvensensecomproductsmotion-

tracking9-axismpu-9150[40] ADXL345 2017 httpwwwanalogcomenproductsmems

accelerometersadxl345html[41] Arduino Pro Mini 2017 httpswwwarduinoccenMain

ArduinoBoardProMini[42] Bluetooth HC-05 2017 httpwwwthaieasyeleccomdownloads

EFDV390EFDV390_Datasheetpdf[43] B Najafi K Aminian A Paraschiv-Ionescu F Loew C J Bula

and P Robert ldquoAmbulatory system for human motion analysisusing a kinematic sensor monitoring of daily physical activityin the elderlyrdquo IEEE Transactions on Biomedical Engineeringvol 50 no 6 pp 711ndash723 2003

[44] K T Wyne ldquoA comprehensive review of tremorrdquo JAAPAvol 18 no 2 pp 43ndash50 2005

[45] S Gottlieb and O C J Lippold ldquo+e 4-6 Hz tremor duringsustained contraction in normal human subjectsrdquo Journal ofPhysiology vol 336 no 1 pp 499ndash509 1983

[46] M C Barroso G P Esteves T P Nunes L M G SilvaA C D Faria and P L Melo ldquoA telemedicine instrument forremote evaluation of tremor design and initial applications infatigue and patients with Parkinsonrsquos diseaserdquo BiomedicalEngineering OnLine vol 10 no 1 p 14 2011

[47] Android Developers Sensor Manager 2018 httpsdeveloperandroidcomreferenceandroidhardwareSensorManagerhtml

[48] T Deshimaru and Y Ikemi Zen et Self-Control Kairos S AEditions Retz Barcelona Spain 6th edition 1985

[49] K Sekida Zen Training Kairos S A Weatherhill Inc NewYork NY USA 6th edition 2016

[50] D Villalba What Is Zen Miraguano Ediciones MadridSpain 2005 ISBN84-7813-286-4

[51] K Noro T Naruse R Lueder N Nao-i and M KozawaldquoApplication of Zen sitting principles to microscopic surgeryseatingrdquoApplied Ergonomics vol 43 no 2 pp 308ndash319 2012

[52] K Ishihara K Dake T Kashihara and S Ishihara ldquoAn attemptto improve the sitting posture of children in the classroomsrdquo inProceedings of the International MultiConference of Engineersand Computer Scientists vol 3 Hong Kong March 2010

Journal of Healthcare Engineering 9

Data Availability

+e data used in this article are available upon request fromthe first author

Conflicts of Interest

+e authors declare no conflicts of interest

Authorsrsquo Contributions

Inmaculada Plaza and Carlos T Medrano conceived anddesigned the experiments Victor H Rodriguez developedthe measuring kit Victor H Rodriguez and InmaculadaPlaza performed the experiments Victor H Rodriguezanalyzed the data Carlos T Medrano contributed analysistools Inmaculada Plaza Victor H Rodriguez and CarlosT Medrano wrote the paper

Acknowledgments

+e authors would like to thank all volunteers who partici-pated in the study +anks are due to ldquoCONACYT-Gobiernodel Estado de Durango Mexicordquo for the research-grant ref-erence 330795386043+is workwas supported by the ldquoFondoSocial Europeordquo the ldquoGobierno de Aragonrdquo the Ministry ofEconomy and Competitiveness (Project reference TEC2013-50049-EXP) and the University of Zaragoza (Project referenceUZ2017-TEC-02)

References

[1] J Kabat-Zinn ldquoAn outpatient program in behavioral medi-cine for chronic pain patients based on the practice ofmindfulness meditation theoretical considerations and pre-liminary resultsrdquo General Hospital Psychiatry vol 4 no 1pp 33ndash47 1982

[2] J Kabat-Zinn ldquoMindfulness-based interventions in contextpast present and futurerdquo Clinical Psychology Science andPractice vol 10 pp 144ndash156 2003

[3] R A Baer ldquoMindfulness training as clinical interventiona conceptual and empirical reviewrdquo Clinical PsychologyScience and Practice vol 10 pp 125ndash143 2003

[4] P Grossman L Niemann S Schmidt and H WalachldquoMindfulness-based stress reduction and health benefits Ameta-analysisrdquo Journal of Psychosomatic Research vol 57 no 1pp 5ndash43 2004

[5] P Sedlmeier M Eberth D Schwarz et al ldquo+e psychologicaleffects of meditation a meta-analysisrdquo Psychological Bulletinvol 138 no 6 pp 1139ndash1171 2012

[6] M Goyal S Singh N Gould et al ldquoMeditation programs forpsychological stress and well-being a systematic review andmeta-analysisrdquo JAMA Internal Medicine vol 174 no 3pp 357ndash368 2014

[7] B Khoury B Knauper F Pagnini N Trent A Chiesa andK Carriere ldquoEmbodied mindfulnessrdquo Mindfulness vol 8no 5 pp 1ndash12 2017

[8] R Aloneftis ldquo+e role of embodiment in third wave mind-fulness based cognitive therapiesrdquo PsyPAG Quarterly vol 102pp 40ndash43 2017

[9] J M Burg T Probst T Heidenreich and J Michalak ldquoDe-velopment and psychometric evaluation of the bodymindfulnessquestionnairerdquo Mindfulness vol 8 no 3 pp 807ndash818 2017

[10] A Cebolla L Galiana D Campos et al ldquoHow does mind-fulness work Exploring a theoretical model using samples ofmeditators and non-meditatorsrdquoMindfulness pp 1ndash11 2017

[11] S Alidina Mindfulness for Dummies John Wiley amp SonsHoboken NJ USA 2014

[12] E ShoninW Van Gordon andM D Griffiths ldquoPractical tipsfor using mindfulness in general practicerdquo British Journal ofGeneral Practice vol 64 no 624 pp 368-369 2014

[13] S R Bishop M Lau S Shapiro et al ldquoMindfulness a pro-posed operational definitionrdquo Clinical Psychology Science andPractice vol 11 no 3 pp 230ndash241 2004

[14] J Kabat-Zinn ldquoSitting meditationsrdquoMindfulness vol 7 no 6pp 1441ndash1444 2016

[15] C E Jones An Investigation into the Role of Body Posture inMindfulness Practice Doctoral dissertation Canterbury ChristChurch University Canterbury UK 2016

[16] T N Hanh5eMiracle of Mindfulness Beacon Boston MAUSA 1975

[17] C P Cook-CottoneMindfulness and Yoga for Self-RegulationA Primer for Mental Health Professionals Springer PublishingCompany New York NY USA 1st edition 2015

[18] P Kaul J Passafiume R C Sargent and B F OrsquoHaraldquoMeditation acutely improves psychomotor vigilance andmay decrease sleep needrdquo Behavioral and Brain Functionsvol 6 no 1 p 1 2010

[19] A Ahani M Wahbeh M Miller et al ldquoChange in physio-logical signals during mindfulness meditationrdquo in Proceedingsof the IEEE EMBS Conference Osaka Japan July 2013

[20] J J Arch and M G Craske ldquoMechanisms of mindfulnessemotion regulation following a focused breathing inductionrdquoBehaviour Research and 5erapy vol 44 no 12 pp 1849ndash1858 2006

[21] J Vidyarthi and B E Riecke ldquoInteractively mediating ex-periences of mindfulness meditationrdquo International Journal ofHuman-Computer Studies vol 72 no 8 pp 674ndash688 2014

[22] W Y Wong and M S Wong ldquoTrunk posture monitoringwith inertial sensorsrdquo European Spine Journal vol 17 no 5pp 743ndash753 2008

[23] M Dozza L Chiari and F B Horak ldquoA portable audio-biofeedback system to improve postural controlrdquo in Pro-ceedings of the 26th Annual International Conference of the

Table 8 ANOVA results for σa (p values)

Mauchlyrsquos test of sphericitySeat 0239

Sensor location 0000Seatlowast sensor location 0000

Pillairsquos traceSeat 0675

Sensor location 0000Seatlowast sensor location 0 726

Sphericity assumedSeat 0744

Sensor location 0000Seatlowast sensor location 0725

Table 9 ANOVA results for the sway (Axz and Axy versus seat)(p value)

Axz Axy

Mauchlyrsquos test of sphericity 0000 0037Pillairsquos trace 0196 0002Sphericity assumed 0299 0006

8 Journal of Healthcare Engineering

IEEE Engineering in Medicine and Biology Society pp 4799ndash4802 San Francisco CA USA September 2004

[24] S L Whitney J L Roche G F Marchetti et al ldquoA com-parison of accelerometry and center of pressure measuresduring computerized dynamic posturography a measure ofbalancerdquo Gait and Posture vol 33 no 4 pp 594ndash599 2011

[25] W Maetzler M Mancini I Liepelt-Scarfone et al ldquoImpairedtrunk stability in individuals at high risk for Parkinsonrsquosdiseaserdquo PLoS One vol 7 no 3 article e32240 pp 1ndash6 2012

[26] J Houtan R Changbum T Ahn and L Stentz ldquoFall riskanalysis of construction workers using inertial measurementunits validating the usefulness of the postural stability metricsin constructionrdquo Safety Science vol 84 pp 161ndash170 2016

[27] S J Ozinga and J L Alberts ldquoQuantification of posturalstability in older adults using mobile technologyrdquo Experi-mental Brain Research vol 232 no 12 pp 3861ndash3972 2014

[28] S R Lord C Sherrington and H B Menz Falls in OlderPeople Risk Factors and Strategies for Prevention CambridgeUniversity Press Cambridge UK 2nd edition 2007

[29] K Chang S Chen and H Lee ldquoWireless accelerator basedbody posture stability detection and application for medita-tion practitionersrdquo in Proceedings of the International Con-ference on Biomedical and Health Informatics BHIpp 112ndash115 Hong Kong China January 2012

[30] K Chang S Chen H Lee et al ldquoA wireless accelerometer-based body posture stability detection system and its appli-cation for meditation practitionersrdquo Sensors vol 12 no 12pp 17620ndash17632 2012

[31] K M Chang Y T Chun S H Chen et al ldquo+e evaluation ofphysical stillness with wearable chest and arm accelerometerduring Chan Ding practicerdquo Sensors vol 16 p 1126 2016

[32] P M Deshmukh C M Russell L E Lucarino andS N Robinovitch ldquoEnhancing clinical measures of posturalstability with wearable sensorsrdquo in Proceedings of the 34th AnnualInternational Conference of the IEEE EMBS pp 4521ndash4524 SanDiego CA USA September 2012

[33] R P Hubble G A Naughton P A Silburn and M H ColeldquoWearable sensor use for assessing standing balance andwalking stability in people with Parkinsonrsquos disease a system-atic reviewrdquo PloS One vol 10 no 4 article e0123705 2015

[34] Lumo Web Page of Lumo Lift product 2017 httpswwwlumobodytechcomlumoliftutm_expid70259375-21Yz80XOgnSXu9lXiNTep1yA1amputm_referrerhttps3A2F2Fwwwlumobodytechcom2Fscience-of-lumo-run2F

[35] Upright Web Page of Personal Posture Trainer-UprightProduct 2017 httpwwwuprightposecom

[36] Alex Web Page of Alex Product 2017 httpalexposturecom[37] Jins-Meme Web Page of Jins Meme Glasses 2017 httpsjins-

memecomen[38] PGrossman ldquoMindfulness for psychologists paying kind attention

to the perceptiblerdquo Mindfulness vol 1 no 2 pp 87ndash97 2010[39] IMU9150 2017 httpswwwinvensensecomproductsmotion-

tracking9-axismpu-9150[40] ADXL345 2017 httpwwwanalogcomenproductsmems

accelerometersadxl345html[41] Arduino Pro Mini 2017 httpswwwarduinoccenMain

ArduinoBoardProMini[42] Bluetooth HC-05 2017 httpwwwthaieasyeleccomdownloads

EFDV390EFDV390_Datasheetpdf[43] B Najafi K Aminian A Paraschiv-Ionescu F Loew C J Bula

and P Robert ldquoAmbulatory system for human motion analysisusing a kinematic sensor monitoring of daily physical activityin the elderlyrdquo IEEE Transactions on Biomedical Engineeringvol 50 no 6 pp 711ndash723 2003

[44] K T Wyne ldquoA comprehensive review of tremorrdquo JAAPAvol 18 no 2 pp 43ndash50 2005

[45] S Gottlieb and O C J Lippold ldquo+e 4-6 Hz tremor duringsustained contraction in normal human subjectsrdquo Journal ofPhysiology vol 336 no 1 pp 499ndash509 1983

[46] M C Barroso G P Esteves T P Nunes L M G SilvaA C D Faria and P L Melo ldquoA telemedicine instrument forremote evaluation of tremor design and initial applications infatigue and patients with Parkinsonrsquos diseaserdquo BiomedicalEngineering OnLine vol 10 no 1 p 14 2011

[47] Android Developers Sensor Manager 2018 httpsdeveloperandroidcomreferenceandroidhardwareSensorManagerhtml

[48] T Deshimaru and Y Ikemi Zen et Self-Control Kairos S AEditions Retz Barcelona Spain 6th edition 1985

[49] K Sekida Zen Training Kairos S A Weatherhill Inc NewYork NY USA 6th edition 2016

[50] D Villalba What Is Zen Miraguano Ediciones MadridSpain 2005 ISBN84-7813-286-4

[51] K Noro T Naruse R Lueder N Nao-i and M KozawaldquoApplication of Zen sitting principles to microscopic surgeryseatingrdquoApplied Ergonomics vol 43 no 2 pp 308ndash319 2012

[52] K Ishihara K Dake T Kashihara and S Ishihara ldquoAn attemptto improve the sitting posture of children in the classroomsrdquo inProceedings of the International MultiConference of Engineersand Computer Scientists vol 3 Hong Kong March 2010

Journal of Healthcare Engineering 9

IEEE Engineering in Medicine and Biology Society pp 4799ndash4802 San Francisco CA USA September 2004

[24] S L Whitney J L Roche G F Marchetti et al ldquoA com-parison of accelerometry and center of pressure measuresduring computerized dynamic posturography a measure ofbalancerdquo Gait and Posture vol 33 no 4 pp 594ndash599 2011

[25] W Maetzler M Mancini I Liepelt-Scarfone et al ldquoImpairedtrunk stability in individuals at high risk for Parkinsonrsquosdiseaserdquo PLoS One vol 7 no 3 article e32240 pp 1ndash6 2012

[26] J Houtan R Changbum T Ahn and L Stentz ldquoFall riskanalysis of construction workers using inertial measurementunits validating the usefulness of the postural stability metricsin constructionrdquo Safety Science vol 84 pp 161ndash170 2016

[27] S J Ozinga and J L Alberts ldquoQuantification of posturalstability in older adults using mobile technologyrdquo Experi-mental Brain Research vol 232 no 12 pp 3861ndash3972 2014

[28] S R Lord C Sherrington and H B Menz Falls in OlderPeople Risk Factors and Strategies for Prevention CambridgeUniversity Press Cambridge UK 2nd edition 2007

[29] K Chang S Chen and H Lee ldquoWireless accelerator basedbody posture stability detection and application for medita-tion practitionersrdquo in Proceedings of the International Con-ference on Biomedical and Health Informatics BHIpp 112ndash115 Hong Kong China January 2012

[30] K Chang S Chen H Lee et al ldquoA wireless accelerometer-based body posture stability detection system and its appli-cation for meditation practitionersrdquo Sensors vol 12 no 12pp 17620ndash17632 2012

[31] K M Chang Y T Chun S H Chen et al ldquo+e evaluation ofphysical stillness with wearable chest and arm accelerometerduring Chan Ding practicerdquo Sensors vol 16 p 1126 2016

[32] P M Deshmukh C M Russell L E Lucarino andS N Robinovitch ldquoEnhancing clinical measures of posturalstability with wearable sensorsrdquo in Proceedings of the 34th AnnualInternational Conference of the IEEE EMBS pp 4521ndash4524 SanDiego CA USA September 2012

[33] R P Hubble G A Naughton P A Silburn and M H ColeldquoWearable sensor use for assessing standing balance andwalking stability in people with Parkinsonrsquos disease a system-atic reviewrdquo PloS One vol 10 no 4 article e0123705 2015

[34] Lumo Web Page of Lumo Lift product 2017 httpswwwlumobodytechcomlumoliftutm_expid70259375-21Yz80XOgnSXu9lXiNTep1yA1amputm_referrerhttps3A2F2Fwwwlumobodytechcom2Fscience-of-lumo-run2F

[35] Upright Web Page of Personal Posture Trainer-UprightProduct 2017 httpwwwuprightposecom

[36] Alex Web Page of Alex Product 2017 httpalexposturecom[37] Jins-Meme Web Page of Jins Meme Glasses 2017 httpsjins-

memecomen[38] PGrossman ldquoMindfulness for psychologists paying kind attention

to the perceptiblerdquo Mindfulness vol 1 no 2 pp 87ndash97 2010[39] IMU9150 2017 httpswwwinvensensecomproductsmotion-

tracking9-axismpu-9150[40] ADXL345 2017 httpwwwanalogcomenproductsmems

accelerometersadxl345html[41] Arduino Pro Mini 2017 httpswwwarduinoccenMain

ArduinoBoardProMini[42] Bluetooth HC-05 2017 httpwwwthaieasyeleccomdownloads

EFDV390EFDV390_Datasheetpdf[43] B Najafi K Aminian A Paraschiv-Ionescu F Loew C J Bula

and P Robert ldquoAmbulatory system for human motion analysisusing a kinematic sensor monitoring of daily physical activityin the elderlyrdquo IEEE Transactions on Biomedical Engineeringvol 50 no 6 pp 711ndash723 2003

[44] K T Wyne ldquoA comprehensive review of tremorrdquo JAAPAvol 18 no 2 pp 43ndash50 2005

[45] S Gottlieb and O C J Lippold ldquo+e 4-6 Hz tremor duringsustained contraction in normal human subjectsrdquo Journal ofPhysiology vol 336 no 1 pp 499ndash509 1983

[46] M C Barroso G P Esteves T P Nunes L M G SilvaA C D Faria and P L Melo ldquoA telemedicine instrument forremote evaluation of tremor design and initial applications infatigue and patients with Parkinsonrsquos diseaserdquo BiomedicalEngineering OnLine vol 10 no 1 p 14 2011

[47] Android Developers Sensor Manager 2018 httpsdeveloperandroidcomreferenceandroidhardwareSensorManagerhtml

[48] T Deshimaru and Y Ikemi Zen et Self-Control Kairos S AEditions Retz Barcelona Spain 6th edition 1985

[49] K Sekida Zen Training Kairos S A Weatherhill Inc NewYork NY USA 6th edition 2016

[50] D Villalba What Is Zen Miraguano Ediciones MadridSpain 2005 ISBN84-7813-286-4

[51] K Noro T Naruse R Lueder N Nao-i and M KozawaldquoApplication of Zen sitting principles to microscopic surgeryseatingrdquoApplied Ergonomics vol 43 no 2 pp 308ndash319 2012

[52] K Ishihara K Dake T Kashihara and S Ishihara ldquoAn attemptto improve the sitting posture of children in the classroomsrdquo inProceedings of the International MultiConference of Engineersand Computer Scientists vol 3 Hong Kong March 2010

Journal of Healthcare Engineering 9

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom