14
© Kamla-Raj 2016 Ethno Med, 10(1): 85-98 (2016) Gamma Probe for Revealing Cancerous Cells Anastasia Yagnyukova 1 , Dmitry Mikhaylov 2 , Timur Khabibullin 3 , Andrey Grigorenko 4 and Leonid Panfilov 5 National Research Nuclear University “MEPhI” (Moscow Engineering Physics Institute), Kashirskoe highway 31, 115409, Moscow, Russian Federation E-mail: 1 < [email protected]>, 2 <[email protected]>, 3 <[email protected]>, 4 <[email protected]>, 5 <[email protected]> KEYWORDS Gamma Probe. Gamma Ray Detection. Cancer Detection. Lymph Node. Radiopharmaceutical ABSTRACT This paper deals with the development of the compact wireless gamma probe for malignancies diagnostics. It defines the location of a ‘guardian’ lymph node (the lymph node closest to the tumor) and defines the size of the tumor, helping establish the stage of disease, the radicalism of the forthcoming operation and the subsequent treatment methods. The gamma probe detects gamma radiation from radiopharmaceuticals accumulating in places of concentration of metastasis, thus revealing cancerous lymph nodes. This paper provides a description of the gamma probe prototype. The main characteristics of the gamma probe were measured, and some problems and were discovered, along with possible solutions. Simulation of the gamma probe was run to determine the optimal configuration of the collimator. Some tests on laboratory animals were run, yielding important measurements, which proved the prototype gamma probe able to quickly and accurately locate the areas of the radiopharmaceutical’s densest concentration within the body. Address for correspondence: Dmitry Mikhaylov National Research Nuclear University “MEPhI”, Kashirskoe Highway 31, 115409, Moscow, Russian Federation Telephone: +7 9037548744 E-mail: [email protected] INTRODUCTION Recognition of oncology diseases using ra- dionuclide diagnostics is today the major task of nuclear medicine among the wide range of other tasks (Giammarile 2014; Brouwer 2014; Von Meyenfeldt 2014; Bogliolo 2015). Radionuclide or radioisotope diagnostics is a section of radio diagnosis designated for detecting pathologi- cal changes in human organs and tissues using radiopharmaceuticals that interfere with physi- ological processes in the body. Diagnostic equipment used in radionuclide medicine generally consists of a detector, an analyzing and converting unit, and a storage device. Amongst such devices, the compact gamma locator, or gamma probe, is specifically designed to identify localized areas of accumu- lation of radiopharmaceuticals within the body (Wengenmair 2005; Gamma Probes 2014; Grig- orenko 2015). Gamma probes are mostly used in the detec- tion of intraoperative sentinel lymph nodes (SLN) and non-invasive scanning of the patient to identify superficial malignant neoplasms (Be- lyaev 2010; Ikram 2013; Bellotti 2013; Mihaljevic 2014; Wei 2015). In cancer staging, the SLN pro- cedure is the most common method to assess the stage to which a cancer has progressed (Dusi 2000; Chen 2006; Da Costa 2006; Sarikaya 2008). The first method is carried out as follows. Before the operation, the patient is given radiop- harmaceuticals that accumulate in the very ni- dus of the tumor, and in the network of nearby lymph nodes affected by metastases. The sur- geon removes the tumor and then removes the lymph nodes, which are in turn checked for me- tastasis with a gamma probe. Since the network of lymph nodes in the body is a diverging net- work, a one-by-one check for metastasis is a reliable means by which to establish the extent to which metastasis has spread throughout the body. This procedure can reduce the invasive- ness of the tumor removal procedure and save the maximal amount of healthy tissue without the risk of recurrence. The second method is completely non-inva- sive and is an addition to traditional radionu- clide procedures. In some cases (superficial lo- cation of the tumor, its small diameter) using sin- gle-photon emission computed tomography (SPECT) or positron emission tomography (PET) imaging of the whole body is irrational, since the cost of a single procedure is important. Due to the limited number of scanners per medical

Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

Embed Size (px)

Citation preview

Page 1: Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

© Kamla-Raj 2016 Ethno Med, 10(1): 85-98 (2016)

Gamma Probe for Revealing Cancerous Cells

Anastasia Yagnyukova1, Dmitry Mikhaylov2, Timur Khabibullin3,Andrey Grigorenko4 and Leonid Panfilov5

National Research Nuclear University “MEPhI” (Moscow Engineering Physics Institute),Kashirskoe highway 31, 115409, Moscow, Russian Federation

E-mail: 1< [email protected]>, 2<[email protected]>,3<[email protected]>, 4<[email protected]>, 5<[email protected]>

KEYWORDS Gamma Probe. Gamma Ray Detection. Cancer Detection. Lymph Node. Radiopharmaceutical

ABSTRACT This paper deals with the development of the compact wireless gamma probe for malignanciesdiagnostics. It defines the location of a ‘guardian’ lymph node (the lymph node closest to the tumor) and definesthe size of the tumor, helping establish the stage of disease, the radicalism of the forthcoming operation and thesubsequent treatment methods. The gamma probe detects gamma radiation from radiopharmaceuticals accumulatingin places of concentration of metastasis, thus revealing cancerous lymph nodes. This paper provides a descriptionof the gamma probe prototype. The main characteristics of the gamma probe were measured, and some problemsand were discovered, along with possible solutions. Simulation of the gamma probe was run to determine theoptimal configuration of the collimator. Some tests on laboratory animals were run, yielding important measurements,which proved the prototype gamma probe able to quickly and accurately locate the areas of the radiopharmaceutical’sdensest concentration within the body.

Address for correspondence:Dmitry MikhaylovNational Research Nuclear University “MEPhI”,Kashirskoe Highway 31, 115409,Moscow, Russian FederationTelephone: +7 9037548744E-mail: [email protected]

INTRODUCTION

Recognition of oncology diseases using ra-dionuclide diagnostics is today the major taskof nuclear medicine among the wide range ofother tasks (Giammarile 2014; Brouwer 2014; VonMeyenfeldt 2014; Bogliolo 2015). Radionuclideor radioisotope diagnostics is a section of radiodiagnosis designated for detecting pathologi-cal changes in human organs and tissues usingradiopharmaceuticals that interfere with physi-ological processes in the body.

Diagnostic equipment used in radionuclidemedicine generally consists of a detector, ananalyzing and converting unit, and a storagedevice. Amongst such devices, the compactgamma locator, or gamma probe, is specificallydesigned to identify localized areas of accumu-lation of radiopharmaceuticals within the body(Wengenmair 2005; Gamma Probes 2014; Grig-orenko 2015).

Gamma probes are mostly used in the detec-tion of intraoperative sentinel lymph nodes (SLN)and non-invasive scanning of the patient to

identify superficial malignant neoplasms (Be-lyaev 2010; Ikram 2013; Bellotti 2013; Mihaljevic2014; Wei 2015). In cancer staging, the SLN pro-cedure is the most common method to assessthe stage to which a cancer has progressed (Dusi2000; Chen 2006; Da Costa 2006; Sarikaya 2008).

The first method is carried out as follows.Before the operation, the patient is given radiop-harmaceuticals that accumulate in the very ni-dus of the tumor, and in the network of nearbylymph nodes affected by metastases. The sur-geon removes the tumor and then removes thelymph nodes, which are in turn checked for me-tastasis with a gamma probe. Since the networkof lymph nodes in the body is a diverging net-work, a one-by-one check for metastasis is areliable means by which to establish the extentto which metastasis has spread throughout thebody. This procedure can reduce the invasive-ness of the tumor removal procedure and savethe maximal amount of healthy tissue withoutthe risk of recurrence.

The second method is completely non-inva-sive and is an addition to traditional radionu-clide procedures. In some cases (superficial lo-cation of the tumor, its small diameter) using sin-gle-photon emission computed tomography(SPECT) or positron emission tomography (PET)imaging of the whole body is irrational, sincethe cost of a single procedure is important. Dueto the limited number of scanners per medical

Page 2: Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

86 ANASTASIA YAGNYUKOVA, DMITRY MIKHAYLOV, TIMUR KHABIBULLIN ET AL.

center, their capacity is low (Romanova 2013). Insuch cases, the rational option is to conduct alocal radio diagnostic procedure in the area inproximity to the tumor. The procedure compris-es administering radiopharmaceuticals to thepatient and subsequent scanning of the loca-tion of the malignancy using a portable gammaray detector (Karamat 2015).

Today many scientists focus their researchon probes for cancer detection use. Some devel-opments include non-invasive probes, whileothers focus on the creation of invasive typesof probes. Both types open new opportunitiesfor data collection, which could ease the work ofsurgeons and other medical specialists.

The outstanding results in targeting cancerwith modern positron emitting radiopharmaceu-ticals have inspired the development of a num-ber of handheld beta probes to assist in surger-ies. The development and initial testing of onesuch probe, a miniature beta-sensitive detectorunit, was developed and initially tested by Ray-lman and Hyder (2004). It proved suitable foruse in minimally invasive procedures.

Barber et al. (2007) developed imaging de-vices small enough to be incorporated in medi-cal probes and inserted into the gastrointestinaltract, imaging concentrations of tumor-seekingradiotracers.

There is another interesting piece of workon nuclear probes and their use as radiation de-tectors in surgery to get information about thedistribution of a radioactive labeled structure.The research was conducted by Navab et al.(2008) and his team. They extended the existingnuclear probes with a spatial localization sys-tem in order to generate functional 3D surfaceimages or functional tomographic images thatcan be extremely useful in the operating room.

Another example is two types of miniaturegamma ray endoscopic probes developed forinsertion into a body cavity to detect invisiblysmall or deeply located tumors. One of them is asingle detector system using a small BGO(Bi

4Ge

3O

12) crystal connected with a fiber optic

light guide, and the other is a dual detector sys-tem using a pair of single detector probes of thesame size combined with a random coincidencetechnique. This research is described by Dha-nasopon et al. (2005).

Existing developments show good results.However, they may have a number of disadvan-tages such as accuracy, size, power consump-tion and high cost.

This paper provides results of compact wire-less gamma probe development for radio-diag-nosis of carcinoma. The gamma probe detectsgamma radiation from radiopharmaceuticals thataccumulate in places of metastasis concentra-tion and so reveal cancerous lymph nodes. Thegamma probe will be able to define how deeplymetastasis has penetrated into tissues.

METHODOLOGY

A prototype of a gamma probe for intraoper-ative use following interstitial injection of a ra-diopharmaceutical to locate regional lymphnodes by their radioactivity has been designedand developed. The prototype comprises a de-tecting part and the electronics. The sensitivepart of the detector is made of a scintillationcrystal LaBr3: Ce (5x10 mm3 cylinder) and a pho-to detector SiPM Hamamatsu 3x3 mm2. Electron-ic components are placed on a multilayer print-ed circuit board mounted within the cylindricalplastic body (Fig. 1).

The voltage bias supply for the SiPM andthe discrimination threshold adjustment is byway of a personal computer (PC), which is con-nected to the probe via a USB cable. The output(detector count rate) is displayed on the PCscreen. Special software has to be installed onthe PC to communicate with the gamma probe.

Details of the count rate of the detector dur-ing each measurement are stored in a text file,which allows further processing of the results.

Circuit Configuration

The electronics unit of the gamma probe con-sists of a voltage bias supply for the SiPM and

Fig. 1. Layout of gamma probe prototypeSource: Author

Page 3: Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

GAMMA PROBE FOR REVEALING CANCEROUS CELLS 87

an amplification discrimination circuit of the an-alogue signal detector. The direct voltage trans-ducer DC-DC MAX 1932 (DC is a direct current)is powered with the voltage from the computer’sUSB connector (+5 V). DC output is set from +25to +75 V. Precise adjustment of the bias voltageis achieved by energizing a second SiPM volt-age contact of the same polarity value from 0 to1000 mV. The voltage is set programmaticallyusing a digital to analogue converter (DAC).

The analogue signal from the photo detec-tor is supplied to an amplifier and then to a com-parator on which by means of DAC the desiredreference voltage Uref is set. At the output ofthe comparator information already in digital for-mat is fed to the input of a field programmablegate array (FPGA), which is programed to per-form the necessary operations on the data fromthe detector.

A block diagram of a prototype gamma-probeis shown in Figure 2. The electronics module isconfigured as a compact printed circuit board(Fig. 3). The circuit allows for the mounting of asecond SiPM and organization of the coinci-dence circuit.

It is notable that such a configuration of thegamma probe has a significant drawback, as afixed connection with the PC is required, the pro-totype cannot be tested in the operating room,

and the data output to the monitor is inconve-nient for doctors. Further development of theprobe provides a new prototype responding tothe following recommendations of radiologistsand surgeons (FSBI ‘Medical Radiological Re-search Centre’, Russian Ministry of Health, Ob-ninsk, Russia): Wireless data transmission An additional data output unit (a monitor

for digital display of the detector countrate)

Audible indication of detector count rate Miniaturization of the printed circuit board

and the body of the probe Storing data on a miniature memory card

for further detailed analysisPossibility of signal output both in digital

and in analogue form (for spectrometric detec-tor characteristics studies).

Fig. 2. A block diagram of a prototype gamma probeSource: Author

Fig. 3. Physical configuration of printed circuitboarSource: Author

Detector andSIPM

2 amplifier 2 comparators

UDAC

SPI Bus

Page 4: Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

Collimation

The protection for the sides of the sensitivepart of the gamma radiation provides the conewith a narrow field of view of the detector. More-over, reducing the diameter of the collimator’’shole means improving the spatial resolution ofthe instrument. However, this inevitably leadsto a loss of efficiency of the detector.

A tungsten cylinder with 2 mm wall thick-ness and inner diameter of 5 mm was producedfor the prototype of the gamma probe (Fig.4).

Selection of the Optimal Collimator:Computer Simulation

Several computer simulations of gammaprobe were conducted to determine the optimalconfiguration of the collimator. The Gate v6.1framework was used for Monte Carlo simulationallowing the development of detector systemmodels using Geant4 libraries (2014). Gate v6.1is an advanced open source software dedicatedto numerical simulations in medical imaging andradiotherapy (Gate, 2014). Three types of colli-mators were examined including simple cylindri-cal (Fig. 5), cylindrical with varying holes at theend (Fig.6), as well as a cylinder with a cone atthe end (Fig. 7).

In the case of the collimator of type (1) wallthickness varied, in the case of type, and (2) theradius of the hole at the end varied. In the caseof collimator of type (3) length and inner radiusof the cone varied.

The collimator’s specifications include a lightgrasp, spatial resolution and spatial selectivity.As spatial resolution and selectivity decline whenthe light grasp ratio increases, the purpose ofthe simulation is to choose a configuration with

Fig. 4. Sensitive part of the gamma probe protectedwith a tungsten collimatorSource: Author

Fig. 5. Simple cylindrical collimator (1)Source: Author

Fig. 6. Collimator with a hole at the end (2)Source: Author

Fig. 7. Collimator with a cone at the end (3)Source: Author

88 ANASTASIA YAGNYUKOVA, DMITRY MIKHAYLOV, TIMUR KHABIBULLIN ET AL.

Page 5: Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

GAMMA PROBE FOR REVEALING CANCEROUS CELLS 89

an optimal compromise between all specifica-tions of the system.

It is notable that the priority of each individ-ual criterion depends on the type of tumor. Ta-ble 1 presents a list of some types of tumors,and a “+” denotes the significance of the probe’sparameters (Wengenmair 2005).

Consequently, the choice of the collimatorshould not only take into account the system’sspecification, but also the future scope for theprobe. It is consistent to have a set of inter-changeable collimators, the most suitable foreach individual case.

Description of the Model

Mathematical modeling of the gamma probewas performed using the package Gate v6.1, withsubsequent processing of the data in MatlabR2010b framework (Matlab (2014) is a high-levellanguage and interactive environment for numer-ical computation, visualization, and programing).

The simulation algorithm can be divided intothe following stages:

1. Description of the model in the Gate frame-work

a. Setting the system’s geometry: size andrelative position of components of thedetector and materials.

The model describes a detector consisting ofa cylindrical scintillation crystal LaBr

3:Ce

(radius 2.5 mm, length 10 mm), aluminumcapsules with a scintillator inside (alumi-num thickness 0.2 mm), and a tungstencollimator of different types.

b. Description of the physics of interactionof gamma rays with the detector’s material.

The model describes the physical processesoccurring during the interaction of low-energygamma rays (photoelectric effect, Compton scat-tering, Rayleigh scattering) and electrons (theionization energy losses, radiation losses, mul-tiple scattering) with the matter.

c. Source description.

The gamma-ray source in the model is a Co-57,as it was used to get the experimental data.

d. Simulation of the electronics.The Gate v6.1 package allows the simulation

of the analogue signal processing using a Digi-tizer unit. The discrimination threshold is set on(100 keV), dead time for the system is introduced(~100 ns), and the energy resolution of the de-tector is determined (8% at 662 keV). Values arechosen based on experimental work with thedetector.

e. Definition of the format of the output data.The Gate output data are at the same time

the initial data for a Matlab program. This ex-plains the choice of the ASCII (American Stan-dard Code for Information Interchange) formatfor output information, as it is universal for bothvirtual environments.

f. Setting of the required statistics/datasettime.

When modeling each experiment, the datasettime was determined so that the resultingstatistics provided a fractional error ofno more than a few percent.

2. Data processing in Matlab.a. Reading the output Gate data.b. Building the necessary distributions and

graphs.c. Calculation of the relevant parameters of

the gamma probe.

Modeling: The Effectiveness of Gamma Rays

The efficiency of gamma rays at different dis-tances between the detector and the source wasmeasured in order to verify the mathematicalmodel. The configuration of the collimator fol-lowed that of the prototype, that is, a tungstencylinder with inner radius of 2.5 mm and wallthickness of 2 mm. A series of measurementswas conducted, and the distance between thesource and the detector was changed from 100cm to 250 cm. Co-57 was selected as the simulat-ed gamma ray source. Its activity equaled theactivity of the laboratory source, which provid-

Table 1: Significance of the probe’s parameters according to different types of tumors’

Melanoma Breast cancer Prostate cancer Tumour of neck,talpa

Detection efficiency + + + +Spatial selectivity + + +Spatial resolution + (head, neck, collar bone) +

Page 6: Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

90 ANASTASIA YAGNYUKOVA, DMITRY MIKHAYLOV, TIMUR KHABIBULLIN ET AL.

ed the corresponding experimental data(3*105Bq).

Figure 8 shows the geometry of the model. Apoint source is located on a line passing throughthe axis of the detector. The distance betweenthe source and the detector is measured with 10mm steps. The dataset time for each point is 100seconds.

The measurement results are presented inFigure 9. The solid line in the graph is the calcu-lated number of gamma rays emitted by thesource into the solid angle of the detector pertime unit. It represents one hundred percent de-tection efficiency. Square dots indicate the num-ber of particles registered by the detector in thesimulation, and triangular points correspond tothe experimental data (see part 3 Experiments).

Detection efficiency of gamma rays is deter-mined by the detector’s count rate of the num-ber of particles emitted at a given solid angle perunit/time ratio. It is constant for every value ofthe distance between the source and the detec-tor. Detection efficiency was 73 ± 2 percent.

Detection efficiency depends on the scintil-lator material and remains constant with variousconfigurations of the collimator. Therefore, it isconsistent to determine the light grasp of thedetector in order to compare the collimators.

The light grasp of the detector with a simplecylindrical collimator (1) at the distance of 100mm is 1.05*10-4. Since this type of collimator cor-responds to the maximum fraction of particlesentering the detector (in comparison with othercollimators), this value is taken as one hundredpercent, and the detector’s light grasp with oth-er collimators will be normalized to it.

Modeling Spatial Resolution

Spatial resolution is defined as the minimumdistance between two point sources, with whichthey are individually distinguishable. In the caseof a single point source, the measure of spatialresolution is full width at half maximum of thedependence of the detector’s count rate on thelocation the source on the line perpendicular tothe axis of the detector.

Geometry of the model is shown in Figure 10.

The distance between the detector and thesource is 10 mm, and the dataset time is 10 sec-onds. The measurement result for one of thetypes of collimator (type 1) is shown in Figure 11.

For convenience in defining the half-widthof the histogram, this was approximated by aGaussian distribution (Lehmann 1997). Samplestatistics is ~105 particles in the peak, meaningthat the fractional error of the spatial resolutionis ~0.1 percent.

Simulation: Spatial Selectivity

Spatial selectivity of the gamma probe is de-termined by the slope angle, at which the detec-

Fig. 10. Geometry of the experiment to determinethe spatial resolution of the detectorSource: Author

Fig. 8. Measurement geometry of detectionefficiencySource: Author

Fig. 9. Detector’s counting rate - distance fromthe source relationship; square pixels are model,triangulars are experiment.Source: Author

50

45

40

35

30

25

20

15

10

5

0

N,

S-1

90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260

x, sm

Page 7: Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

GAMMA PROBE FOR REVEALING CANCEROUS CELLS 91

tor must be rotated regarding the source, so thatthe count rate halved.

Measurement geometry is shown in Figure12. Distance from the source to each detector is30 cm, and the angle varies from -90° to 90° witha step of 16°. Dataset time was 100 seconds. Themeasurement result for one of the types of colli-mators (type 1) is shown in Figure 13.

The histogram is approximated by a Gauss-ian distribution for the convenience of calcula-tion of the half width.

Statistics of about 103 particles providesa fractional measurement error of about fivepercent.

The Simulation Results

Measurements of light grasp, spatial resolu-tion and spatial selectivity were taken for threetypes of collimators, namely, open cylinder, thecylinder with a hole in the front end and thecylinder with a conical front part. In the firstcase, the cylinder wall thickness varied (fromlimiting case, without a collimator, to a thicknessof 2 mm). In the second type of the collimatorthe radius in the front end varied from 2.5 mm(open end) to 1 mm. In the case of a conical endof the collimator, the inner radius of the coneand its length varied.

The simulation results are shown in Table 2.The table shows that the use of tungsten

collimator with a wall thickness more than 1 mmis inconsistent, but the tungsten cylinder with athin wall is fragile and difficult to manufactureand operate.

As a conclusion, it can be noted that theoptimum configuration is a cylinder with a coneat the end, since it allows the light grasp valueto be kept at almost the same level as in the caseof open collimator, while greatly improving thespatial resolution, and is the only collimatorwhich it is able to improve spatial selectivity.

RESULTS

The developed prototype was tested. Mea-surements of the main characteristics of gammaprobe, such as spatial resolution, spatial selec-tivity and the efficiency of detection of gammarays were taken.

Laboratory gamma ray source of Cs-137 (662keV) and Co-57 (80% of 122 keV and 136 keV,10%) was used. All measurements were madeusing a tungsten collimator with inner radius of2.5 mm and wall thickness of 2 mm.

Integral Spectrum of Cs-137

The prototype gamma probe is unable tooutput the analogue signal from the detectingpart to the standard spectrometric tract, but vi-

Fig. 12. Measurement geometry of the experimentto determine the spatial selectivity of the detectorSource: Author

Fig. 13. Angular distribution of detector count rateSource: Author

Fig. 11. Coordinate distribution of the detector’scount rateSource: Author

1000

900

800

700

600

500

400

300

200

100

0

N

-100 -80 -60 -40 -20 0 20 40 60 80 100Theta, grad

Page 8: Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

92 ANASTASIA YAGNYUKOVA, DMITRY MIKHAYLOV, TIMUR KHABIBULLIN ET AL.

sualization of the energy spectrum is very im-portant in terms of properly selecting the dis-crimination limit and evaluating the general ac-curacy of the detector.

It is known that the integral and differentialenergy spectra are in a one-to-one relation, thatis, the integral spectrum is a histogram of pulsesfrom the detector, having amplitude above thethreshold. Therefore, values of each point ofthe integral of the spectrum can be obtained bysummation of the differential spectrum to theleft of the threshold.

Thus, there is a special feature in the area ofthe photopeak in both differential spectrum andintegrated spectrum. This fact was verified bymeans of a package of Matlab R2010b. An ex-pected integral spectrum was graphed (Fig. 14b) based on the previous measurements in theenergy spectrum of the source of Cs-137 (Fig. 14a). The figures clearly state that in the region ofthe left edge of the photopeak on the integralspectrum there is a sloping area in the differen-tial spectrum. Thus, using a gamma probe of theintegrated spectrum and setting a discrimina-tion threshold in the middle of the sloping areaallows the cutting off of events associated withCompton scattering.

Experimental data proved that a sloping areais present in the integral spectrum of the sourceof Cs-137 (Fig. 15, highlighted in dashed lines).

The integral spectrum was obtained by re-cording the detector count rate as a function ofthe discrimination threshold programmaticallyexposed.

Table 2: Simulation of a probe with different types of collimators (compared to the light-grasp of thedetector with an open cylinder collimator (Type 1))

Collimator type Wall Inner Outer Length, Light- Spatial Spatialthickness, radius, radius, mm grasp, resolu- selecti-m m m m m m %* tion, mm vity, °

Cylinder No 2.5 - 10 100 12.5 -collimator0.5 3 100 8.2 271 3.5 100 7.8 251.5 4 100 7.8 252 4.5 100 7.8 25

Cylinder with a closed end 2 1 4.5 12 19 2.4 191.5 37 4.5 192 68 5.2 192.5 95 6.9 19

Cylinder with a cone 2 1 4.5 15 18 4.5 13 at the end 1.5 34 5.4 13

2 67 6.9 132.5 95 7.8 13

2 2.5 4.5 12 96 14.4 23.613 96 10.6 1614 96 9.9 15.915 96 7.8 13

Fig. 14. a. Differential spectrum of Cs-137; b. In-tegral spectrum, based on differential spectrum.Source: Author

6 0 0

5 0 0

4 0 0

3 0 0

2 0 0

1 0 0

0

0 100 200 300 400 500 600 700 800 900 Chemical

100 200 300 400 500 600 700 800 900

Threshold

2

1 .5

1

0 .5

x 105

a

b

NN

, in

t

Page 9: Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

GAMMA PROBE FOR REVEALING CANCEROUS CELLS 93

SiPM Temperature Instability

During the measurements a decrease in thedetector count rate over time (Fig. 16) was dis-covered.

Investigation of the problem revealed thatthe decrease in the count rate is due to the ther-mal instability of the silicon photomultiplier inoperating mode the photodiode heats and SiPMgain falls with increasing temperatures, and thetemperature increase necessitates a correspond-ing increase in the value of the operating biasvoltage that must energize SiPM to keep the gain.

Special measurements were taken in a ther-mostat and an obtained experimental plan wasgraphed in order to define the nature of the rela-tionship between the working voltage and tem-perature measurements. Figure 17 shows thatthe dependence is linear. The index is 56 mV/°Ñ.

There are two options for the temperaturecompensation of bias voltage, and both are

based on the inclusion of a thermistor in thecharge. In one of them, changing the thermistorresistance value is digitized and converted intocode, which is then input to the DAC and setsthe necessary additional voltage on the secondSiPM contact. The second method is analogue,which is a thermistor and is included in the feed-back circuit, provided by developers of the volt-age converter MAX 1932.

Figure 18 (MAX1932 2007) shows a typicalcircuit with the NTC-thermistor (negative temper-ature coefficient) in a voltage conversion circuit.

The analogue method of thermal stabiliza-tion was implemented. The required ratio of volt-age change per degree Celsius bias is provided

Fig. 15. The integral spectrum of the source ofCs-137 (662 keV)Source: Author

Fig. 16. Counting rate – time relationshipSource: Author

Fig. 17. Operating voltage bias – working tem-perature SiPM dependenceSource: Author

Fig. 18. Schematic diagram of the inclusion ofthe thermistorSource: Author

4 0 0

3 5 0

3 0 0

2 5 0

2 0 0

1 5 0

1 0 0

5 0

0

500 700 900 1100 1300 1500Threshold, Mv

pps

1 8

1 6

1 4

1 2

1 0

8

6

4

2

0

N,

S-1

0 20 40 60

Time, min

73 .1

73 .0

72 .9

72 .8

72 .7

72 .6

72 .5

72 .4

72 .3

72 .2

U.V

20 22 24 26 28 30 32 34 36

T, 0C

VIN

TOCS+ TOCS-

VOUT

GATE

MAX 1932FB

R5

R3

R9R3NTCTHERMISTOR

Page 10: Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

94 ANASTASIA YAGNYUKOVA, DMITRY MIKHAYLOV, TIMUR KHABIBULLIN ET AL.

by selecting the rating value for resistors R5,R8, R9 in Figure 18. First, operation of the pro-posed scheme was verified using a multi-turntrimmer resistor. Manual changing of its resis-tance simulated the thermistor. The expected lin-ear dependence of the voltage and ‘imaginarytemperature’ was obtained.

Further measurements were taken directlyfrom the thermistor. The measurement result isshown in Figure 19. The graph shows the out-put voltage compensation of circuit MAX 1932on the temperature, and the dependence is lin-ear with the compensation factor of 40 mV/ °C.

Gamma Ray Detection Efficiency

Measurements of gamma ray detection effi-ciency were taken on a Co-57 source, as its line(124 keV) is close in energy to the line of themost common medical radioisotope Tc-99m (140keV) and Cs-137 (662 keV). These measurementshave the effectiveness of a single geometry (onlyone distance between source and detector), butat several points were taken in order to test theresults. Efficiency value must be maintained atdifferent distances between the source and thedetector.

The measurement results for the Co-57 andCs-137 are shown in Figure 20 (up and downrespectively).

The calculated curves of the detector countrate and the distance from the source depen-dence, in the approximation of a point source atone hundred percent efficiency is shown in thegraphs along with the experimental data. Thus,

the efficiency of the detector is the ratio of thecount rate on the experimental curve to the cal-culated value of the counting rate at a givendistance.

The effective detection of Gamma ray was 70± 2 percent for Co-57 (124 keV) and 29 ± 4 per-cent for the Cs-137 (662 keV). These values areconstant over the whole range of distancesshown in the graph, for both sources.

Experimental points are well approximated bythe dependence of 1/R2 type, and hence in thepresent range of the point source’s distances,approximation is valid.

Sensitivity of the detector was also mea-sured. It is determined as the number of pulsesper second recorded by the detector, referred toone kBq of source activity measured at the veryend of a sensitive part of the locator. This valuereached 12 c/s/kBq at the Co-57 source.

Fig. 19. Compensation of the voltage bias using aNTC-thermistorSource: Author

73.0

72 .9

72 .8

72 .7

72 .6

72 .5

72 .4

72 .3

72 .2

72 .1

U.V

18 20 22 24 26 28 30 32 34 36 38

T, 0C

5 0

4 5

4 0

3 5

3 0

2 5

2 0

1 5

1 0

0

80 100 120 140 160 180 200 220 240 260X, mm

N,

S-1

N,

S-1

a

b 2 6

2 4

2 2

2 0

1 8

1 6

1 4

1 2

1 0

8

6

4

2

0

70 80 90 100 110 120 140 160 170 180X, mm

Fig. 20. The dependence of counting rate of thedetector on distance from the source, a. for Co-57, b. for Cs-137Source: Author

Page 11: Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

GAMMA PROBE FOR REVEALING CANCEROUS CELLS 95

Spatial Resolution and Spatial Selectivity

Spatial resolution describes the minimum dis-tance between two point sources on which theycan be registered separately. The dependencyof the detector count rate for the source posi-tion along a line perpendicular to the axis of thedetector should be measured in order to deter-mine the spatial resolution (Fig. 21). Full widthat half maximum of the distribution (FWHM) isthe estimation of the spatial resolution.

The inner diameter of the scintillator is 5 mm,and the wall thickness of the tungsten collima-tor is 2 mm.

Projection distance between the detector andthe source of the detector onto the axis is select-ed to be 1 cm, since this is a typical arrangementof nodes at the depth of the human body.

Spatial selectivity is the angular resolutionof the detector. Spatial selectivity is determinedas detector count rate dependence on the polarangle between the detector axis and the line con-necting the source and the detector (Fig. 22).FWHM of this distribution is the estimation ofspatial selectivity.

The measurement results are shown in Fig-ure 23 (spatial resolution, SR) and Figure 24 (spa-tial selectivity, SS). FWHM value was 8 mm and26°, respectively. All measurements were takenwith a Co-57 source (124 keV).

Laboratory Tests on Animals

The prototype gamma probe was tested on alaboratory rat, which was given “Tehnefor”(99mTc-EDTMP) radiopharmaceutical. The medi-cation is bone-tropic and accumulates mainly inthe bones of the skeleton. About forty percent

Fig. 21. Scheme for spatial resolution measuringSource: Author

Fig. 22. Scheme for spatial selectivity measuringSource: Author

Fig. 23. Dependence of the counting rate on thetransverse coordinate between the detector andthe source. FWHM = 8 mm.Source: Author

Fig. 24. Dependence of the counting rate on thepolar angle between the detector and the source.FWHM = 26°Source: Author

Collimator

Scintillator

1 6 0

1 4 0

1 2 0

1 1 0

1 0 0

9 0

8 0

7 0

6 0

5 0

4 0

3 0

2 0

1 0

0

N,

S-1

-10 -8 -6 -4 -2 0 2 4 6 8 10

X, mm

-60 -40 -20 0 20 40 60

4 .5

4 .0

3 .5

3 .0

2 .5

2 .0

1 .5

1 .0

0 .5

0 .0

N,

S-1

θ, 0

Page 12: Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

96 ANASTASIA YAGNYUKOVA, DMITRY MIKHAYLOV, TIMUR KHABIBULLIN ET AL.

of the given dose is excreted by the kidneys. Inthe study, three hours after being administeredmost of the radiopharmaceutical was detectedin the rat’s bladder, which measures ~1 cm. Thisallowed the device to be tested in conditionsclose to its real application areas.

Figure 25 (left) shows a diagram of the rat’sbody scan. A net was put on the rat, and thecount rate was recorded in the net’s nodes. Apicture shown in Figure 25 (right) was obtainedbased on this two-dimensional array of valuesusing Matlab R2010b packet. The Figure showsthat apart from the region of maximum accumu-lation of the radiopharmaceutical (bladder), oth-er soft tissues are also visible, and the contourof the body is clearly portrayed.

Based on the same data a three-dimensionalsurface graph showing the difference in thecount rate in the maximum accumulation area ofthe radiopharmaceutical (in comparison withneighboring areas) was created (Fig. 26).

CONCLUSION

The proposed prototype of the gamma probefor revealing cancerous cells in tissues was com-

pared with analogues, namely, Europrobe CsJ,Eurorad, C-Trak Omni-Probe and Neoprobe. Theresults of measurements of principal parametersof the gamma probe and their comparison withtheir Western analogies are shown in Table 3.

The prototype is no worse than foreign ana-logues. Its spatial resolution and selectivity areparticularly notable as advantages of the gam-ma probe. When in operating conditions, porta-bility and the ability to control the device with asmall PC are of particular importance.

However, more studies with radiopharmaceu-ticals should be conducted, including preclini-cal tests on animals with model diseases andclinical tests on patients.

RECOMMENDATIONS

The proposed gamma probe will significant-ly increase the effectiveness of patient examina-tion and allow detecting metastases in organ-isms. It will help prevent severe forms of onco-logical diseases. Portability of the device andthe opportunity of control with the help of asmall computer play a great role in an operatingroom.

Fig. 25. Photo of the lab rat and radiopharmaceuti-cal accumulation image obtained by gamma probeSource: Author

Table 3: Comparison of parameters of gamma probes

Manufacturer Coordinate Spatial Sensitivity,(country) resolution, mm selectivity, ° imp/s/kBq

Europrobe CsJ, Eurorad (France) 14 35 7C-Trak Omni-Probe, Care Wise (USA) 15 50 23Neoprobe 2000 (USA) 15 36 10Gamma-probe (Russia, NRNU MEPhI) 8 26 12

Fig. 26. Spatial distribution of the counting rateSource: Author

100000

10000

1 0 0 0

1 0 0

N

02

46

81 0

1 2

Y, mm

1 0

2

46

8

1 0

X, mm

Page 13: Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

GAMMA PROBE FOR REVEALING CANCEROUS CELLS 97

In the feature, it is planned to finish the stageadvanced development for the gamma probe,prepare commercial design for the device, andto organize small lot manufacturing and developworking constructing documents.

The device needs clinical testing.

REFERENCES

Barber HB, Barrett HH, Wild WJ, Woolfenden JM 2007.Development of small in vivo imaging probes fortumour detection. IEEE Transactions on NuclearScience, 31(1): 599-604.

Bellotti C, Castagnola G, Tierno SM, Centanini F, Spar-agna A, Vetrone I, Mezzetti G 2013. Radioguidedsurgery with combined use of gamma probe and hand-held gamma camera for treatment of papillary thy-roid cancer locoregional recurrences: a preliminarystudy. European Review for Medical and Pharmaco-logical Sciences, 17(24): 3362-3366.

Belyaev VN Brantova ÒS Emelyanova ÎV 2010. Gam-ma-probe based on a semiconductor photodetectorfor rapid diagnosis of cancer. Medical Physics, 2: 42-47.

Bogliolo S, Marchiole P, Sala P, Giardina E, Villa G,Fulcheri E, Menada MV 2015. Sentinel node map-ping with radiotracer alone in vulvar cancer: A fiveyear single-centre experience and literature review.European Journal of Gynaecological Oncology,36(1):10-15.

Brouwer OR, Van Den Berg NS, Mathéron HM, VanDer Poel HG, Van Rhijn BW, Bex A, Van Tinteren H,Valdés Olmos RA, Van Leeuwen FWB, Horenblas S2014. A hybrid radioactive and fluorescent tracer forsentinel node biopsy in penile carcinoma as a poten-tial replacement for blue dye. European Urology,65(3): 600-609.

Chen SL, Iddings DM, Scheri RP, Bilchik AJ 2006.Lymphatic mapping and sentinel node analysis: Cur-rent concepts and applications. A Cancer Journalfor Clinicians, 56(5): 292-309.

Da Costa FE, Rela PR, de Oliveira IB, Pereira MCC,Hamada MM 2006. Surgical gamma probe with TlBrsemiconductor for identification of sentinel lymphnode. IEEE Transactions on Nuclear Science, 53(3):1403-1407.

Dhanasopon AP, Levin CS, Foudray AMK, Olcott PD,Habte F 2005. Scintillation crystal design featuresfor a miniature gamma ray camera. IEEE Transac-tions on Nuclear Science, 52(5): 1439-1446.

Dusi W, Angelotti P, Bollini D, Mengoni D, Morom C,Ricard M 2000. Advanced gamma-probes for radiogu-ided oncological surgery. IEEE Nuclear Science Sym-posium Conference Record, 3: 21/118-121/123.

Friess H, Martignoni ME 2014. Transferring innova-tive freehand SPECT to the operating room: Firstexperiences with sentinel lymph node biopsy in ma-lignant melanoma. European Journal of SurgicalOncology, 40(1): 42-48.

Gamma Probes 2014. IntraMedical Imaging. From<http://www.gammaprobe.com/products/gamma-probes> (Retrieved on 21 April 2014).

GATE 2014. Forewords. From <http://www. opengate-collaboration. org> (Retrieved on 21 April 2014).

Geant4: A Toolkit for the Simulation of the Passage ofParticles Through Matter. 2014. Last updated: 21Mar 2014. From <http://geant4.web.cern.ch/geant4>(Retrieved on 1 April 2014).

Giammarile F, Bozkurt MF, Cibula D, Pahisa J, OyenWJ, Paredes P, Olmos RV, Sicart SV 2014. The EANMclinical and technical guidelines for lymphoscintig-raphy and sentinel node localization in gynaecolog-ical cancers. European Journal of Nuclear Medicineand Molecular Imaging, 1-15.

Grigorenko Andrey, Panfilov Leonid, Smirnov Alex-ander, Starikovskiy Andrey, Rubin Dmitry, ShulgaEkaterina, Sychev Nikolay, Nikolaeva Alexandra2015. The existing gamma-probes review for search-ing functional increase and complex improvementpossibilities. Biosciences Biotechnology ResearchAsia, 12(Spl. Edn. 2): 197-200.

Ikram M, Akhtar S, Maseeh-uz-Zaman Junaid M, DhariT, Ahmad Z, Hussain R 2013. Sentinel node localisa-tion using pre-operative lymphoscintigraphy andintraoperative gamma probe in early oral cavity can-cer. Journal of the Pakistan Medical Association,63(8): 976-979.

Karamat MI, Farncombe TH 2015. Simultaneous 99mTcand 111In SPECT reconstruction using acceleratedconvolution-based forced detection Monte Carlo.Nuclear Science, IEEE Transactions on, 62(5): 2085-2095.

Lehmann K 1997. Gaussian Distributions. PrincetonUniversity. From <http://www.edasolutions.com/old/MathCad/Statistics/images/GaussianDistribution.pdf>(Retrieved on 1 April 2014).

MATLAB® 2014. The MathWorks, Inc. From <http://www.mathworks.com/products/matlab.> (Retrievedon 14 April 2014)

MAX1932 2007. Digitally Controlled, 0.5% Accurate,Safest APD Bias Supply. Maxim Integrated. Page LastModified: 2007-07-13. From <http://www. maxim-integrated. com/datasheet/index.mvp/id/3484> (Re-trieved on 11 February 2015).

Mihaljevic AL, Rieger A, Belloni B, Hein R, Okur A,Scheidhauer K, Schuster T, Sarikaya I, Sarikaya A,Reba RC 2008. Gamma probes and their use in tu-mour detection in colorectal cancer. InternationalSeminars in Surgical Oncology, 5: 25.

Navab N, Traub J, Buck AK, Ziegler SI, Wendler T2008. Navigated Nuclear Probes for Intra-operativefunctional imaging. Proceedings of 5th IEEE Inter-national Symposium on Biomedical Imaging (ISBI2008), Paris, France, 14-17 May 2008, pp. 1395-1398.

Romanova S 2013. Nuclear medicine: Status and pros-pects. Remedium, 6: 8-20.

Raylman RR, Hyder 2004. A dual surface barrier de-tector unit for beta-sensitive endoscopic probes. IEEETransactions on Nuclear Science, 51(1): 117-122.

Von Meyenfeldt EM, Siebenga J, Van Der Pol HAG,Schreurs WMJ, Hulsewe KWE 2014. Radionuclide-guided biopsy of bone lesions in cancer patients: Areliable, well-tolerated technique. European Jour-nal of Surgical Oncology, 40(2): 193-196.

Wei L, Chen F, Zhang X, Li D, Yao Z, Deng L, Xiao G2015. 99mTc-dextran lymphoscintigraphy can de-tect sentinel lymph node in breast cancer patients.Experimental and Therapeutic Medicine, 9(1): 112-116.

Page 14: Gamma Probe for Revealing Cancerous Cellskrepublishers.com/02-Journals/S-EM/EM-10-0-000-16-Web/S-EM-10-1-16... · Gamma Probe for Revealing Cancerous Cells ... identify superficial

98 ANASTASIA YAGNYUKOVA, DMITRY MIKHAYLOV, TIMUR KHABIBULLIN ET AL.

Wengenmair H, Kopp J 2005. Gamma Probes for Sen-tinel Lymph Node Localization: Quality Criteria,Minimal Requirements and Quality of CommerciallyAvailable Systems. From <http://www.klinikum-

augsburg.de/index.php/fuseaction/download/lrn_file/gammaprobes.pdf> (Retrieved on 5 February 2015).

Paper received for publication on June 2014Paper accepted for publication on November 2015.