BIOPHOTONICS - scholar.harvard.edu · Journal of BIOPHOTONICS 1/17 1/17 Laserthermaltherapymonitoringusingcomplex differentialvarianceinopticalcoherencetomography

Journal of

BIOPHOTONICS1/171/17

www.biophotonics-journal.org

Laser thermal therapy monitoring using complexdifferential variance in optical coherence tomographyW. C. Y. Lo, N. Uribe-Patarroyo, A. S. Nam, M. Villiger,B. J. Vakoc, B. E. Bouma

FULL ARTICLE

Laser thermal therapy monitoring using complexdifferential variance in optical coherence tomography

William C. Y. Lo*, 1, 2, Néstor Uribe-Patarroyo1, Ahhyun S. Nam1, 3, Martin Villiger1,Benjamin J. Vakoc1, 2, and Brett E. Bouma1, 2

1 Wellman Center for Photomedicine and Department of Dermatology, Massachusetts General Hospital and Harvard Medical School,40 Blossom Street, Boston, Massachusetts 02114, USA

2 Harvard-Massachusetts Institute of Technology Division of Health Sciences and Technology, Cambridge, Massachusetts 02142, USA3 Department of Mechanical Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge,Massachusetts 02139, USA

Received 27 March 2016, revised 10 July 2016, accepted 21 August 2016Published online 13 September 2016

Key words: laser thermal therapy, optical coherence tomography, retina, skin, esophagus

1. Introduction

The current state-of-the-art techniques for clinicalthermal therapy monitoring typically involve the useof temperature as part of the treatment feedback loop.Unfortunately, temperature measurements usingthermocouples only permit point sampling, whileemerging magnetic resonance thermometry techni-ques [1, 2] and 15O-water PET imaging [3] offer lowspatiotemporal resolution and are expensive to use.

More recently, photoacoustic thermography of tissuewas also investigated for the assessment of myocardialablation lesions [4]. However, these temperature-based monitoring techniques can only provide an in-direct estimate of tissue injury and the spatial resolu-tion remains low for epithelial applications.

Within the domain of optical coherence tomogra-phy (OCT), several early studies explored the use ofpolarization-sensitive OCT (PS-OCT) for the assess-ment of thermal damage directly [5, 6]. In particular,

* Corresponding author: e-mail: [email protected]

Conventional thermal therapy monitoring techniquesbased on temperature are often invasive, limited by pointsampling, and are indirect measures of tissue injury, whiletechniques such as magnetic resonance and ultrasoundthermometry are limited by their spatial resolution. Thevisualization of the thermal coagulation zone at high spa-tial resolution is particularly critical to the precise deliveryof thermal energy to epithelial lesions. In this work, an in-tegrated thulium laser thermal therapy monitoring systemwas developed based on complex differential variance(CDV), which enables the 2D visualization of the dy-namics of the thermal coagulation process at high spatialand temporal resolution with an optical frequency domainimaging system. With proper calibration to correct fornoise, the CDV-based technique was shown to accuratelydelineate the thermal coagulation zone, which is markedby the transition from high CDV upon heating to a signifi-cantly reduced CDV once the tissue is coagulated, in 3different tissue types ex vivo: skin, retina, and esophagus.

The ability to delineate thermal lesions in multiple tissuetypes at high resolution opens up the possibility of per-forming microscopic image-guided procedures in a vastarray of epithelial applications ranging from dermatology,ophthalmology, to gastroenterology and beyond.

J. Biophotonics 10, No. 1, 84–91 (2017) / DOI 10.1002/jbio.201600072

© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

reduced tissue birefringence was observed with PS-OCT in thermally damaged porcine skin or tendondue to the denaturation of collagen, which is consis-tent with earlier studies in thermally damaged myo-cardium [7] and rat tail tendon [8] using polarizinglight microscopy. PS-OCT was also shown to be use-ful for burn assessment in patients [9, 10] and forstudying the remodeling of hypertrophic scars long-itudinally in vivo [11]. In addition, catheter-basedPS-OCT was demonstrated for radiofrequency abla-tion monitoring in porcine myocardium ex vivo [12],following earlier observation of changes in birefrin-gence bands between untreated and ablated myocar-dial tissue with intensity-based OCT [14–16]. Re-cently, depolarization (in addition to linear retard-ance) was investigated to visualize radiofrequency-ablated myocardial tissue ex vivo using Mueller ma-trix polarimetry in a backscattered geometry [16].However, these techniques are typically limited totissue with high birefringence at baseline (e.g., mus-cle, skin, or tendon) and their broad application toreal-time monitoring of thermal therapy in diversetissue types remains challenging. As an alternativeapproach, we previously demonstrated the feasibilityof visualizing injury depth in laser thermal therapyof porcine esophagus based on phase variations inOCT; however, this requires careful phase calibra-tion or a highly phase-stable OCT system and wasdemonstrated in an M-mode imaging configuration,enabling only 1-D, single-point monitoring withoutstructural information [17].

In the present work, we developed a 2-D thermaltherapy monitoring algorithm (with potential exten-sion to 3-D) using complex differential variance [18],which exploits the phase and intensity fluctuationsduring thermal therapy but does not require a highlyphase-stable OCT system. The algorithm is applied

to the OCT data directly, so visualization of structur-al information simultaneously during therapy moni-toring is not inhibited. To assess the feasibility ofthis approach for monitoring laser thermal therapy,a wavelength tunable thulium fiber laser (1860–1895 nm) was used, which takes advantage of a strongwater absorption peak for thermal coagulation in bio-logical tissue. Using the integrated thulium laser andOCT imaging setup, we demonstrate the ability ofthe CDV-based technique to accurately delineate thethermal coagulation boundary in 3 different tissuetypes: retina, skin, and esophagus, which opens upthe possibility of performing microscopic image-guided interventions at high spatiotemporal resolu-tion in numerous clinical applications.

2. Materials and methods

2.1 Laser thermal therapy experimental setup

Figure 1 shows the experimental setup used for thelaser thermal therapy study. A detailed description ofthe optical frequency domain imaging system was re-ported previously [19]. Briefly, the system features afiber ring wavelength-swept laser operating at240 kHz with 4-fold interleaving, a center wavelengthof 1310 nm and a 125 nm sweep range. An acousto-optic modulator was used in the reference arm to re-move depth-degeneracy [20]. An imaging window of10 mm, consisting of 1024 A-lines/B-scan, wasscanned with a focused beam resulting in a 1/e2 radiusof 32 μm of the lateral point spread function in thesquared norm of the complex-valued tomogram.

A wavelength tunable thulium fiber laser (IPGPhotonics, Oxford, MA), which provides single-

Figure 1 Schematics of the laser thermal therapy monitoring setup integrating an optical frequency domain imaging systemwith a wavelength tunable thulium fiber laser. Simultaneous OCT imaging and laser therapy with a thulium fiber laser isachieved using a dichroic mirror and programmable shutter to synchronize the timing of therapy. AOM, acousto-optic mod-ulator; PC, polarization controller; PD, photodiode; FBG, fiber Bragg grating.

J. Biophotonics 10, No. 1 (2017) 85

© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheimwww.biophotonics-journal.org

mode output from 1860 nm to 1895 nm, was used forlaser thermal therapy. The benchtop setup inte-grated the thulium laser into the OCT optical pathusing a dichroic mirror. The collimated beam with a1/e2 diameter of ~800 μm, measured with the Win-CamD UCD12 beam profiler (DataRay, Redding,CA), was aligned to the center of the imaging planeon the tissue surface and the wavelength was set to1890 nm to match the water absorption peak. Outputpower was varied from 300 mW to 500 mW (60 W/cm2 to 100 W/cm2) to investigate the technique atvarious fluence rates, including those reaching tissuevaporization threshold [21].

2.2 Processing algorithm

Figure 2 provides an overview of the processing al-gorithm developed for laser thermal therapy moni-toring and coagulation zone visualization in thisstudy. We employed complex differential variance(CDV) [18], which exploits both intensity and phasevariations, to map the dynamic fluctuations in thecomplex OCT signal during the thermal coagulationprocess and extended this implementation to cali-brate for noise and to enable the intuitive visualiza-tion of the coagulation zone. For a pair of A-linesacquired at times t and t + 1 (in M consecutive B-scans), CDV is computed as follows:

Figure 2 Overview of processing al-gorithm for laser thermal therapymonitoring and coagulation zone vi-sualization. Complex B-scans areprocessed in batches ofM B-scans toproduce raw CDV (CDVraw) frames.Using a calibration window in a staticregion within the tissue, a calibrationcurve for the SNR dependence is es-timated and applied to produce thecalibrated CDV frames (CDVcal).The final image is produced by over-laying the intensity image on top (I-CDVraw vs. I-CDVcal), showing sig-nificant improvements with calibra-tion. The coagulation zone is visua-lized by detecting regions that havereached the cumulative CDV thresh-old (ξ) and returned to a low instan-taneous CDVupon coagulation (ε).

fCDV zð Þ ¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

1�PM�1

t¼0

��PL

l¼�Lw lð Þ R z� l; tð Þ R� z� l; t þ 1ð Þ

��PM�1

t¼0

PLl¼�L

wðlÞ 12 ½ R z� l; tð Þj j2 þ R z� l; t þ 1ð Þj j2�

vuuuuuut ð1Þ

where R(z, t) is the complex OCT signal, R*(z, t)its complex conjugate, w(k) is a depth window func-tion with length 2L + 1, and M indicates the num-ber of consecutive B-scans used for averaging.Here, a Hanning window with a length of 11 (L= 5)depth pixels was used and M was set to 10 B-scans(which results in an effective frame rate of 23 fpswith 1024 A-lines/B-scan at a 240 kHz A-line rate).While this algorithm intrinsically rejects phase noiseinduced by small axial bulk tissue motion and syn-chronization errors, we observed an SNR-depend-ent increase in CDV in static regions resulting in anon-zero baseline CDV in deeper regions due todecreased SNR. Hence, we performed an SNR-de-pendent calibration of the CDV using a static re-gion within the tissue next to the therapy zone (ca-libration window size: 20 × 10 pixels after down-sampling by 10× in both the intensity image andraw CDV image, each with an original size of 1024× 1024 pixels). The SNR-dependent CDV was fittedto an exponential decay and subtracted from theraw CDV to derive the calibrated CDV values(CDVcal). The calibrated CDV was mapped ontothe intensity OCT images for simultaneous visuali-zation of underlying structures. Finally, the coagula-tion zone was visualized by highlighting regions thathave reached a cumulative CDV threshold ξ and re-turned to a low CDV below ε upon coagulation,where the cumulative CDV as a function of time t

W. C. Y. Lo et al.: Laser thermal therapy monitoring using complex differential variance86

© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.biophotonics-journal.org

is defined as

cCDVcal tð Þ ¼Xt

τ¼0

CDVcal τð Þ ð2Þ

and the coagulation zone c(t), which is a binarymask that evolves as a function of time t, is deter-mined as follows:

c tð Þ ¼ cCDVcal tð Þ > ξf g \ CDVcal tð Þ < εf g ð3Þ

2.3 Histological analysis

Tissue samples (bovine eye, porcine skin, and por-cine esophagus) were harvested immediately aftersacrifice and laser thermal therapy was performed exvivo at room temperature. Tissue marking dyes wereplaced on each side of the laser therapy spot for co-registration of OCT images with histology. Tissuesamples were trimmed and sectioned for either rou-tine H&E histology or nitroblue tetrazolium chloride(NBTC) histology (with eosin counterstain) afterembedding in optimum cutting temperature com-pound for frozen sections at 10 μm. NBTC stains forthe activity of NADH-diaphorase, which subsidesupon thermal coagulation and cell death [22].

3. Results

Figure 3 shows an example of monitoring laser ther-apy in bovine retina ex vivo using the CDV-basedapproach, demonstrating the ability to clearly visua-lize the dynamics of the laser-induced heating andcoagulation process. The center therapy zone exhi-bits a rapid rise in CDV (purple) extending from theretina into the sclera upon the initiation of lasertherapy at 300 mW (60 W/cm2). A growing zonewith reduced CDV (yellow), surrounded by a rim ofhigh CDV (purple/red), was observed as the processcontinued. This growing zone (marked by the bluedotted lines on each image) corresponds well histo-logically to the thermal coagulation zone marked bythe denaturation of the collagen in the fibrous scleralayer in H&E.

Figure 4 shows a similar thermal coagulation sig-nature in the rapidly expanding zone with reducedCDV surrounded by high CDV for the case of por-cine skin ex vivo at the same power setting of300 mW (60 W/cm2). The panel clearly reveals theearly dynamics of the coagulation process for thefirst 1.5 s. Lesion formation appeared slightly fasterin this case compared to the retina, most likely dueto differences in tissue properties and the presence

Figure 3 Laser thermal therapy monitoring in bovine retinaex vivo using calibrated complex differential variance(CDVcal). Images represent cross-sectional slices at the cen-ter of the thulium laser therapy beam (300 mW or 60 W/cm2) every 2 s for 10 s, where CDVcal was overlaid on inten-sity images. H&E histology (inset) reveals a coagulationzone extending into the sclera that corresponds with the re-gion with reduced CDV (yellow) at the center delineatedby an expanding boundary with high CDV (purple). Color-bar (CDVcal) ranges from 0 to 0.5. Blue dotted lines, coagu-lation zone. r, retina; c, choroid; s, sclera. Scale bars= 500 μm.

Figure 4 Laser therapy monitoring in porcine skin ex vivoat 300 mW (60 W/cm2) for 1.5 s. NBTC histology with eosincounterstaining (inset) at treatment endpoint confirms thatthe extent and shape of thermal damage in the dermis(NBTC-negative region in pink) corresponds well with theexpanding zone with reduced CDV (yellow) delineated bya boundary with high CDV (purple). Colorbar (CDVcal)ranges from 0 to 0.5. Blue dotted lines, coagulation zone. e,epidermis; d, dermis. Asterisks (*), air bubbles. Scale bars= 500 μm.



of a small layer of vitreous humor on top of the reti-na which absorbed part of the thermal energy deliv-ered. NBTC histology was performed to confirm theboundary of thermal coagulation, and the NBTC-ne-gative region, indicative of thermal damage, corre-sponds well with the coagulation zone indicated bythe CDV-based technique.

To further investigate this approach for delineat-ing lesion boundaries in different tissue types and atdifferent laser power settings, we performed anotherexperiment in porcine esophagus ex vivo using laserpower from 300 mW to 500 mW, corresponding tofluence rates of 60 W/cm2 to 100 W/cm2 (Figure 5).The lesion boundary was particularly well defined atthe early phase (t = 1–3 s), but became less clear asthe thermal coagulation zone expanded, especially atthe lowest fluence rate (60 W/cm2). Interestingly, atthe highest fluence rate tested (100 W/cm2), a super-ficial layer with increased CDV was observed insidethe therapy zone due to the formation of expandingair bubbles that eventually burst with tissue vaporiza-tion (at t = 4 s when the instantaneous CDV reachedthe maximum value) as confirmed by histology.

To better visualize the evolution of the coagulationzone over time, the history of CDV values was used todetect regions that have reached the coagulationthreshold as defined in Eq. (3), highlighted in purpleand overlaid onto the intensity images (Figure 6). Thisvisualization technique overcomes the limitation ofdisplaying the instantaneous CDV, when the bound-ary becomes less visible with the expansion of the coa-gulation zone, especially at the lower power settings.Even at the highest power setting tested (when the tis-sue vaporization threshold was reached), this ap-proach still provided a good estimate of the coagula-tion zone, as confirmed by NBTC histology.

Finally, histological validation with NBTC stain-ing was performed at representative time points dur-ing laser therapy from 1 s to 5 s at 400 mW (80 W/cm2), showing that the coagulation zone delineatedby the CDV approach corresponds well histologi-cally to the NBTC-negative region and thermal da-mage (Figure 7). In particular, at the early phase oftherapy (t = 1 s), the CDV-based technique indicatesan injury depth limited to the epithelium (reachingthe lamina propria at t = 3 s) while at the treatment

Figure 5 Effect of increasing the output power of the thulium fiber laser during laser therapy in porcine esophagus ex vivo:(a) 300 mW [60 W/cm2], (b) 400 mW [80 W/cm2] (Supplementary Video 1), and (c) 500 mW [100 W/cm2]. NBTC histologywith counterstaining was performed at the treatment endpoint (t = 5 s), showing close correspondence between the NBTC-negative region (pink) and the zone with reduced CDV (yellow) delineated by a boundary with high CDV (purple). Color-bar (CDVcal) ranges from 0 to 0.5. Asterisks (*), air bubbles. Scale bars = 500 μm.

Figure 6 Visualization of the coagulation zone in porcine esophagus using the history of CDV values at different powersettings: (a) 300 mW [60 W/cm2], (b) 400 mW [80 W/cm2] (Supplementary Video 1), and (c) 500 mW [100 W/cm2]. Purplecolor indicates regions that have reached the coagulation threshold defined by the cumulative CDV parameter and lowinstantaneous CDV (ξ = 2 and ε = 0.1), which provides an alternative way to more clearly delineate the thermal coagulationzone as confirmed by NBTC histology. Scale bars = 500 μm.



endpoint (t = 5 s), the injury depth has reached themuscularis mucosa, as confirmed histologically byNBTC staining. By contrast, the lesion boundary isdifficult to delineate with the intensity OCT imagesalone despite the increased backscattering observedin the coagulated tissue, especially as the shadow be-low the coagulated zone expanded at later timepoints (Figure 7c).

To further validate the findings systematicallyand quantitatively, three lesions were created foreach therapy duration, and the depth of the coagula-tion zone (injury depth) was measured on corre-sponding CDV images and NBTC histology inde-pendently by three experts, showing good agreementbetween the CDV-based and NBTC-based measure-ments overall (r2 = 0.95) (Figure 8). In cases wherediscrepancies were observed, the depth measured onNBTC histology was typically smaller than thatmeasured using the CDV image (~50–100 μm), mostlikely due to variability in tissue shrinkage duringfrozen section processing as well as instantaneousthermal expansion and bubble formation in thesuperficial layer during therapy.

4. Discussion

In this study, we demonstrated the use of dynamicOCT measurements based on complex differentialvariance to accurately delineate the coagulation zonein laser thermal therapy and validated this approachhistologically in 3 different tissue types: skin, retina,and esophagus. By overlaying this information di-rectly on intensity OCT images, the active treatmentzone was visualized simultaneously with the underly-ing tissue architecture, which enables lesions to betargeted more precisely. Compared to previous workbased on PS-OCT or polarimetry-based techniques[5, 6, 12, 13], the current method is not limited tobiological tissues with high birefringence at baseline(e.g, skin, muscle, and tendon) and has broader im-plications in a number of clinical applications.

However, there are several limitations to the cur-rent CDV-based approach. First, unlike PS-OCTwhere the thermal lesion can be readily imaged andvisualized post treatment due to the loss of tissue bi-refringence, the CDV-based approach derives con-trast from the real-time variations in phase and in-tensity signals that subside once the therapy beam isturned off. The need for real-time tracking alsomeans that motion can lead to an artificial increasein CDV, which can be mitigated with increased imag-ing speed or similarly by decreasing the number ofB-scans (M) used to compute CDV in Eq. (1), aswell as various motion correction techniques [23–25]. Similarly, perfusion in living tissue can contri-bute to the CDV signal, but considering the timescale, the relative contribution will likely be smallerthan the therapy-induced CDV signal. In addition,the active treatment zone boundary can becomemore difficult to delineate as the coagulation zonegrows bigger and the CDV signal degrades as afunction of depth. When the uncalibrated CDV val-ues begin to saturate at depths approaching thenoise floor (typically at a depth >1–1.5 mm), the dy-

Figure 7 Histological validation of CDV-based coagulationzone monitoring in porcine esophagus ex vivo at 400 mW(80 W/cm2) for 5 s. NBTC histology with counterstainingwas obtained at each representative time point (1 s, 3 s, and5 s), showing close correspondence between the NBTC-ne-gative coagulation zone (pink) (a) and the region showingthe transition from high to low CDV (purple to yellow) (b),which is difficult to delineate clearly using intensity imagesalone (c). e, epithelium; lp, lamina propria; mm, muscularismucosa; sm, submucosa. Scale bars = 500 μm.

200 300 400 500 600 700 800 900Injury depth on CDV image (µm)

200

300

400

500

600

700

800

900

Inju

ry d

epth

on

NB

TC h

isto

logy

(µm

)

Figure 8 Quantitative validation of coagulation zone depth(injury depth) in porcine esophagus after laser therapy at400 mW (80 W/cm2) for 1 s, 3 s, and 5 s. Each data pointrepresents the mean ± SD of the injury depths on the OCT(CDV) image and corresponding NBTC histology mea-sured independently by three experts. The solid line repre-sents perfect correlation of OCT and histology measure-ments for comparison. Blue circles, 400 mW (80 W/cm2) at1 s. Red triangles, 400 mW (80 W/cm2) at 3 s. Yellow dia-monds, 400 mW (80 W/cm2) at 5 s.



namics of the heating process becomes increasinglydifficult to observe, especially at lower fluence rates.Therefore, we have performed an SNR-dependentcalibration to enable more robust visualization atgreater depths, which provides an accurate estimateof the lesion boundary at the different depths tested.However, the depth limitation is also a fundamentallimitation of OCT imaging, which is difficult to over-come even with calibration for noise, due to inherentlimits on system sensitivity and penetration depth oflight. To further improve the visualization of thecoagulation zone, we developed an alternative visua-lization method in which the history of CDV valuesis tracked and regions that have reached a cumula-tive CDV threshold ξ and returned to an instanta-neous CDV below ε as defined in Eq. (3) are high-lighted. While the coagulation threshold is likely dif-ferent between tissue types, this visualization strat-egy provides a good estimate of the growingcoagulation zone that would be more difficult to ap-preciate with the instantaneous CDV alone.

Compared to our earlier work on monitoring la-ser therapy with phase variations in M-mode ima-ging [17], the current CDV-based approach was de-veloped to overcome small axial bulk motion andphase jitter from data acquisition synchronizationthrough the use of a depth window as defined in Eq.(1), which eliminates the need for a highly phase-stable OCT system or precise phase calibration [21].We further demonstrated 2-D, cross-sectional ther-apy monitoring in different tissue types (with poten-tial extension to 3-D) to enable simultaneous visuali-zation of underlying tissue structures. It is interestingto observe in both our earlier study and the currentstudy that the active treatment zone is marked by in-creased phase variation or CDV and the growingcoagulation zone corresponds very well to the regionwhere the phase variation or CDV has subsided.

By providing real-time therapy monitoring usingthe CDV-based technique, more targeted and perso-nalized thermal therapy delivery can be achieved indermatology, ophthalmology, gastroenterology, andbeyond. An interesting opportunity provided by thispromising therapy monitoring approach is the abilityto precisely control various laser parameters, such aspower, wavelength, and duration of therapy, to bet-ter tailor the therapy to the target lesion with im-mediate feedback [20]. For example, the currenttechnology can be applied to tailor the laser treat-ment of hypertrophic scars [26], laser photocoagula-tion in diabetic retinopathy especially in the sub-threshold regime [27], and more targeted thermaltherapy of epithelial lesions such as Barrett’s eso-phagus with high-grade dysplasia [17, 20, 25]. Usingthe current approach, more precise thermal therapyin epithelial lesions may be achieved by tailoring thethermal injury to the desired depth; for example, inthe treatment of Barrett’s esophagus, it is important

to avoid excessive damage beyond the submucosa toreduce the risk of stricture formation [29]. The abil-ity to delineate thermal lesions in different tissuetypes, at high spatiotemporal resolution, opens upthe possibility of performing microscopic image-guided procedures in a vast array of epithelial appli-cations in the future.

Supporting Information

Additional supporting information may be found inthe online version of this article at the publisher’swebsite.

Supplementary Video 1: Laser thermal therapymonitoring in porcine esophagus ex vivo using com-plex differential variance (400 mW or 80 W/cm2 for5 s).

Acknowledgements Research reported in this publicationwas funded in part by the National Institutes of HealthP41 EB015903 grant. WCYL was supported by the Cana-dian Institutes of Health Research (CIHR) Doctoral For-eign Study Award.

Author Biographies Please see Supporting Informationonline.

References

[1] O. Seror, M. Lepetit-Coiffé, B. Le Bail, B. D. de Sen-neville, H. Trillaud, C. Moonen, and B. Quesson, Eur.Radiol. 18(2), 408–16 (2008).

[2] B. D. de Senneville, C. Mougenot, B. Quesson, I. Dra-gonu, N. Grenier, and C. T. W. Moonen, Eur. Radiol.17(9), 2401–2410 (2007).

[3] A. Bao, B. Goins, G. D. Dodd, A. Soundararajan, C.Santoyo, R. A. Otto, M. D. Davis, and W. T. Phillips,J. Nucl. Med. 49(10), 1723–1729 (2008).

[4] H. Ke, S. Tai, and L. V. Wang, J. Biomed. Opt., 19(2),26003 (2014).

[5] J. De Boer, S. Srinivas, A. Malekafzali, Z. Chen, andJ. Nelson, Opt. Express 3(6), 212–218 (1998).

[6] K. Schoenenberger, B. W. Colston, D. J. Maitland, L.B. Da Silva, and M. J. Everett, Appl. Opt. 37(25),6026 (1998).

[7] S. Thomsen, J. A. Pearce, and W. F. Cheong, IEEETrans. Biomed. Eng. 36(12), 1174–1179 (1989).

[8] D. J. Maitland and J. T. Walsh, Lasers Surg. Med.20(3), 310–318 (1997).

[9] M. C. Pierce, R. L. Sheridan, B. Hyle Park, B. Cense,and J. F. De Boer, Burns 30(6), 511–517 (2004).

[10] K. H. Kim, M. C. Pierce, G. Maguluri, B. H. Park, M.Lydon, and J. F. De Boer, J. Biomed. Opt. 17(6),066012 (2012).

[11] W. C. Y. Lo, M. Villiger, A. Golberg, G. F. Broelsch,S. Khan, C. G. Lian, W. G. Austen Jr., M. Yarmush,



and B. E. Bouma, J. Invest. Dermatol. 136(1), 84–92(2016).

[12] X. Fu, Z. Wang, H. Wang, Y. T. Wang, M. W. Jenkins,and A. M. Rollins, Opt. Lett. 39(17), 5066–5069(2014).

[13] C. P. Fleming, K. J. Quan, and A. M. Rollins, J.Biomed. Opt. 15(4), 041510 (2013).

[14] C. P. Fleming, H. Wang, K. J. Quan, and A. M. Roll-ins, J. Biomed. Opt. 15(3), 030516 (2013).

[15] C. P. Fleming, K. J. Quan, H. Wang, G. Amit, and A.M. Rollins, Opt. Express 18(3), 3079–3092 (2010).

[16] I. Ahmad, A. Gribble, M. Ikram, M. Pop, and A. Vit-kin, J. Biophotonics 10, 1–10 (2015).

[17] B. J. Vakoc, G. J. Tearney, and B. E. Bouma, J.Biomed. Opt. 12(2), 020501 (2007).

[18] A. S. Nam, I. Chico-Calero, B. J. Vakoc, Biomed.Opt. Express 5(11), 3822–3832 (2014).

[19] C. Jun, M. Villiger, W. Y. Oh, and B. E. Bouma, Opt.Express 22(21), 2919–2921 (2014).

[20] S. Yun, G. Tearney, J. de Boer, and B. Bouma, Opt.Express 12(20), 4822–4828 (2004).

[21] M. Villiger, A. Soroka, G. J. Tearney, B. E. Bouma,and B. J. Vakoc, J. Biomed. Opt. 16(11), 118001–118001 (2011).

[22] R. A. Neumann, R. M. Knobler, F. Pieczkowski, andW. Gebhart, J. Am. Acad. Dermatol. 25(6), 991–998(1991).

[23] N. D. Shemonski, S. S. Ahn, Y.-Z. Liu, F. A. South, P.S. Carney, and S. A. Boppart, Biomed. Opt. Express 5(12), 4131–43 (2014).

[24] V. X. Yang, M. L. Gordon, A. Mok, Y. Zhao, Z.Chen, R. S. Cobbold, B. C. Wilson, and I. Alex Vitkin,Optics Communications 208(4)–(6), 209–214 (2002).

[25] L. An, H. M. Subhush, D. J. Wilson, and R. K. Wang,J. Biomed. Opt. 15(2), 026011 (2010).

[26] R. R. Anderson, M. B. Donelan, C. Hivnor, E. Gree-son, E. V. Ross, P. R. Shumaker, N. S. Uebelhoer, andJ. S. Waibel, JAMA dermatology 150(2), 187–193(2014).

[27] P. Romero-Aroca, J. Reyes-Torres, M. Baget-Bernal-diz, and C. Blasco-Suñe, Curr. Diabetes Rev. 10(2),100–112 (2014).

[28] K. K. Wang and R. E. Sampliner, Am. J. Gastroenter-ol. 103(3), 788–797 (2008).

[29] N. J. Shaheen and D. J. Frantz, Curr. Opin. Gastroen-terol. 26(4), 361–366 (2010).



Documents

BIOPHOTONICS - scholar.harvard.edu · Journal of BIOPHOTONICS 1/17 1/17 Laserthermaltherapymonitoringusingcomplex differentialvarianceinopticalcoherencetomography