Nihms116835

Ultrasound Guidance and Monitoring of Laser-Based Fat Removal

Jignesh Shah, MS1, Sharon Thomsen, MD1,2, Thomas E. Milner, PhD1, and Stanislav Y.Emelianov, PhD1,*1Department of Biomedical Engineering, University of Texas at Austin, Austin, Texas 787122Pathology for Physicists and Engineers, Sequim, Washington 98382

AbstractBackground and Objectives—We report on a study to investigate feasibility of utilizingultrasound imaging to guide laser removal of subcutaneous fat. Ultrasound imaging can be used toidentify the tissue composition and to monitor the temperature increase in response to laserirradiation.

Study Design/Materials and Methods—Laser heating was performed on ex vivo porcinesubcutaneous fat through the overlying skin using a continuous wave laser operating at 1,210 nmoptical wavelength. Ultrasound images were recorded using a 10 MHz linear array-based ultrasoundimaging system.

Results—Ultrasound imaging was utilized to differentiate between water-based and lipid-basedregions within the porcine tissue and to identify the dermis-fat junction. Temperature maps duringthe laser exposure in the skin and fatty tissue layers were computed.

Conclusions—Results of our study demonstrate the potential of using ultrasound imaging to guidelaser fat removal.

Keywordslaser therapy; ultrasound imaging; thermal imaging; treatment monitoring; fat removal; bodyreshaping

INTRODUCTIONLiposuction, also known as lipoplasty, is an invasive procedure for subcutaneous fat removaland body reshaping usually performed under local anesthesia [1]. Recent innovations inliposuction, including ultrasound and laser-assisted liposuction where fat is emulsified beforeapplying suction [1–4], have lead to shorter treatment times and reduced scarring. Despite theseadvances, several disadvantages associated with liposuction are recognized such as scarring,skin sagging, and risk of mortality [1,5,6]. Laser-based treatment for body sculpting or fatremoval is a recently proposed non-invasive alternative to liposuction [7].

Selective laser heating can be achieved by utilizing an optical wavelength where the absorptionby the target tissue is greater than the surrounding region [8]. Specifically for fat treatments,the absorption of lipids at vibration bands near 915, 1,210, and 1,720 nm exceeds that of water

© 2008 Wiley-Liss, Inc.*Correspondence to: Stanislav Y. Emelianov, PhD, Department of Biomedical Engineering, University of Texas at Austin, Austin, TX78712. E-mail: E-mail: [email protected].

NIH Public AccessAuthor ManuscriptLasers Surg Med. Author manuscript; available in PMC 2009 July 22.

Published in final edited form as:Lasers Surg Med. 2008 December ; 40(10): 680–687. doi:10.1002/lsm.20726.

NIH

-PA Author Manuscript

NIH


NIH


[7]. Using 1,210 nm optical wavelength, temperature increases of > 20°C were obtained andfat damage has been demonstrated through the overlying skin [7].

Prior to initiating laser therapy, knowledge of the laser dosimetry required to heat and removethe adipose tissue is required. However, the dermal-fat boundary can vary in depth from 0.5to 4 mm while subcutaneous fat can have a thickness of a few centimeters [9]. Knowledge ofthe tissue composition and depth of the dermis-fat interface is useful information in selectinglaser dosimetry (incident fluence, irradiation wavelength, pulse duration, and exposure time).

The rupture of adipocytes has been observed in response to laser irradiation [2,3]. Themechanism leading to fat breakdown is dependent on both the heating time and the temperatureincrease [10,11]. During laser heating, nonspecific thermal damage to the surrounding tissueis possible and may lead to scarring. For efficacious laser treatment, protecting surroundingtissue structures is essential while ensuring damage to target tissues. A need is recognized fora diagnostic imaging technique to identify the tissue composition before laser therapy andmonitor the depth-resolved temperature increase during therapy.

Ultrasound is a real-time, non-invasive imaging modality that is typically employed in thediagnosis of tissue abnormalities and identification of pathological tissue [12,13]. Ultrasoundimaging has also been utilized for tissue characterization based on temperature dependentchanges of the speed of sound in tissue [14,15]. In addition, ultrasound imaging has beenrecently proposed to monitor the temperature increase in response to laser irradiation [16–18].

We report on experiments to test the feasibility of using ultrasound imaging to guide laser fatremoval. A laboratory setup consisting of an ultrasound imaging system interfaced with acontinuous wave laser was assembled. Experiments were performed on ex vivo porcinesubcutaneous fat through the overlying epidermis and dermis. Results of the experimentsprovide data to test the feasibility of using ultrasound imaging both to identify the dermis-fatjunction and to monitor the temperature increase during therapy.

THEORYUltrasound imaging has been used to monitor temperature changes by measuring the thermallyinduced change in the speed of sound [16–20]. Herein, we present a similar approach adoptedto identify the tissue composition along with measurement of the temperature increase inresponse to laser irradiation outlined in Figure 1. The time-of-flight for ultrasound pulse-echoin a homogenous medium is given by

(1)

where t(T0) is the time delay between the transmitted pulse and an echo from a scatterer atdepth z at initial temperature of T0, and c(T0) is the speed of sound in the medium. When thetemperature rises by ΔT, there are two separate effects—the speed of sound changes and thetissue volume (v(T)) will change due to thermal expansion/ contraction. An apparent time shiftin arrival of the ultrasound signal is observed due to the combined effects of thermal expansionand speed of sound change. The time-of-flight for the ultrasound signal in a heated volume canbe written as

(2)

Shah et al. Page 2

Lasers Surg Med. Author manuscript; available in PMC 2009 July 22.

NIH


NIH


NIH


where α is the linear coefficient of thermal expansion in the specimen, and c(T0+ΔT) is thespeed of sound after the temperature increase. For temperatures below 55°C in tissue, the effectof thermal expansion on the time shift is negligible compared to the speed of sound change[21,22]. The temperature-induced apparent time shift (Δt) of the ultrasound signal can beexpressed as

(3)

Note that the apparent time shift can also be referred to as an apparent spatial shift (since, asindicated in Eq 1, distance and time are equivalent).

From Eq. (3), the apparent time shift depends on speed of sound change and depth/time.Therefore, for the same temperature rise, ultrasound signals from deeper structures will havegreater time shifts. However, this spatial dependence of the time shift is removed bydifferentiating along the axial direction. The term d(Δt(z))/dt, representing the temporal (and,therefore, spatial) gradient of the apparent time shift, is referred to as the normalized time shift.

In water-bearing tissue, such as muscle or skin, the speed of sound increases with a rise intemperature [23]. On the other hand in lipid-based tissues, such as fat, the speed of sounddecreases with a rise in temperature [23]. For example, the speed of sound in bovine liverincreases with temperature at a rate of 1.83 m/(s °C) which is comparable to that of water at2.6 m/(s °C) [24]. In contrast, speed of sound decreases in bovine fat at −7.4 m/(s °C). Sincethe temperature-dependent speed of sound varies significantly between different tissue types,ultrasound-based methods for tissue characterization are possible based on generalcomposition of water-based and lipid-based tissues [14,15]. Specifically, by tracking the timeshifts in ultrasound signal arrival (which is the result of temperature-induced change in thespeed of sound), subdermal fat and water/collagen rich dermis can potentially be differentiatedwith high contrast.

The effective temperature change can be related to the apparent time shift by the followingexpression

(4)

where Δt(z) is the apparent time shift between two ultrasound signals and k is a materialdependent coefficient that can be experimentally determined [18–20]. A temperature controlledwater-bath experiment is typically performed to estimate k, where the relationship betweentemperature rise and normalized time shift is determined [18,20]. Therefore during laserheating, the normalized time shift between successive ultrasound B-scan frames can becomputed and the spatial distribution of the temperature elevation can be determined using Eq.(4) and independently measured or a priori known relationship between temperature rise andnormalized time shift (i.e., coefficient k).

A procedure is envisioned whereby a small laser induced temperature increase is produced andultrasound imaging is used to identify regions of water-based or lipid-bearing tissue regions(Eq. 3 and Eq. 4). A tissue composition map of subdermal structures can be generated bydemarcating the boundary between skin and fat. During laser exposure, ultrasound imagingcan be applied to estimate the spatial distribution of temperature (Eq. 4) in tissue.

Shah et al. Page 3


NIH


NIH


NIH


MATERIALS AND METHODSExperimental Setup

An imaging and therapeutic experimental setup was designed and assembled to acquireultrasound frames during laser irradiation. The diagram of the experimental setup is presentedin Figure 2a and a photograph of the assembly is shown in Figure 2b. The ultrasound signalswere captured using a 128 element linear transducer array operating at a 10 MHz centerfrequency. A continuous wave laser operating at 1,210 nm optical wavelength is utilized fordelivering the radiant energy.

During the laser heating, ultrasound signals were acquired every 0.1 seconds and stored foroffline processing (Fig. 2c). The received signals were then used to reconstruct a grayscale B-mode image using a conventional delay-and-sum beamforming approach.

Tissue PreparationFresh ex vivo porcine tissue samples (15 mm × 15 mm × 12 mm) were obtained with skin andfat intact. The tissue samples were selected having at least 8 mm thickness of subcutaneousfat. The tissue specimen was placed on the holder (Fig. 2a) with the epidermal side on contactwith a sapphire sphere of 3 mm diameter. The laser irradiation was delivered via a 300 µmdiameter fiber to a sapphire sphere, which acts as a focusing lens. The ultrasound transducerwas placed inline with the laser fiber gently touching the adipose-side of the tissue specimen(Fig. 2a).

The experiments were performed at room temperature of 20°C. Prior to the laser irradiation,the tissue samples, which were stored in a refrigerator, were allowed to equilibrate for at least30 minutes. The laser irradiation was applied for 5 seconds with a beam power of 0.9 Wmeasured at the output of the fiber.

Immediately after laser irradiation, the tissue samples were bisected with a blade and fixed informalin. Routine hematoxylin and eosin (H&E) staining was performed on the tissue slicesalong the laser exposure and imaging plane and observed under a light microscope.

Estimation of the Relationship Between Normalized Time Shift and TemperaturePrior to laser irradiation, the temperature response of the porcine tissue was determined usinga temperature controlled water bath experiment. Separate tissue specimens from the sameanimal were placed inside a constant temperature water bath. The temperature of the waterbath was increased from room temperature of 20–55°C in discrete increments. At eachincrement, temperature was maintained constant for 30 minutes. Then, the temperaturedistribution was assumed to be spatially homogenous and an ultrasound frame was recorded.

Normalized time shifts were computed between successive ultrasound frames from two distinctregions in the sample—fatty tissue and skin. Therefore, the relationship between normalizedtime shift and temperature for the porcine fat and skin was measured and approximated usinga second-order polynomial fit to determine the coefficient k in Eq. (4) needed for directestimation of temperature. The normalized time shift decreases for fatty tissue (Fig. 3a) andincreases for non-fatty tissue (Fig. 3b) with temperature. Furthermore, the normalized timeshift changes by a greater amount for fat (∼8%) compared to skin (∼3%) for the sametemperature range. These results are consistent with literature data where the speed of soundfor fat decreases while the speed of sound in water-based tissue increases with temperature[23,24].

Shah et al. Page 4


NIH


NIH


NIH


Data AnalysisWhile performing laser heating, apparent time shifts between successive ultrasound frameswere estimated by employing a correlation-based block matching algorithm [25]. Then theapparent time shifts (Eq. 3) were differentiated along the axial direction to estimate thenormalized time shifts (Eq. 4). The tissue composition map was then generated by identifyingthe sign of the normalized time shift—negative sign indicating fat and positive sign signifyingdermis. Once the tissue composition was computed and the dermal-fat junction wasdetermined, the temperature increase was estimated by applying the coefficients of thepolynomial fit (Fig. 3a,b) to the measured normalized time shift.

RESULTSTissue Boundary Map

The ultrasound image of the ex vivo porcine tissue sample is presented in Figure 4a representinga 10 mm × 15 mm field of view. Note bottom of the ultrasound image is masked to removereverberations of ultrasound pulse from the tissue holder.

Maps of the normalized time shift were generated from successive ultrasound frames duringa 5-second laser exposure. Figure 4b plots the normalized time shift along the region indicatedby the arrows in Figure 4a and represents an axial line from a depth of 3–7 mm from the topof the ultrasound image. The normalized time shifts have two distinct regions. At the 5–7 mmdepth, the normalized time shift is increasing with the laser irradiation time while thenormalized time shift is decreasing otherwise. Since the normalized time shift for porcine skinhas positive temperature gradient while fat has a negative temperature gradient (Fig. 3), theregion exhibiting a positive normalized time shift is classified as skin while the region havingnegative normalized time shift is classified as fat. Furthermore, the zero crossing between thepositive and negative normalized time shift represents position of the dermis-fat junction. Notethat location of the zero-crossing does not change regardless of the laser exposure time, thatis, the increasing magnitude of the normalized time shift does not affect position of the zero-crossing (Fig. 4b).

The map of normalized time shift after the 5 seconds of laser irradiation is shown in Figure 5a.Two distinct regions are visible on the normalized time shift image above the laser irradiationspot (shown by the arrow)—a brighter region having a positive normalized time shift and darkerregion having a negative normalized time shift corresponding to positive and negativetemperature gradients of the normalized time shift. About 5 mm to the left from the laserirradiation spot there is a region with a negative normalized time shift—this area had an imagingartifact produced by the tissue holder and, therefore, was excluded from a subsequent analysis.

The zero-crossing of the normalized time shift, computed over the region corresponding tolaser irradiation, is overlaid on the ultrasound image (Fig. 5b). Overall, the zero-crossingdelineates two regions, the upper region is classified as fat and the brighter region located belowis classified as skin. In this way, the zero-crossing of the normalized time shift may be used toidentify the dermis-fat junction.

Tissue Temperature MapThe temperature map immediately after 5 seconds laser irradiation is shown in Figure 6a. Thetissue composition was first identified using the dermis-fat junction from Figure 5b. Thenormalized time shifts were converted to temperature by using the respective relationshipsestablished for porcine fat (Fig. 3a) and dermis (Fig. 3b). The overlaid map (Fig. 6b) indicatesthat temperature increases by more than 25°C in both skin and fatty tissue regions.

Shah et al. Page 5


NIH


NIH


NIH


In addition, the spatial-temporal temperature rise was examined in four 0.5 mm × 0.5 mmregions. The regions are centered along the laser irradiation direction at depths of 7 mm, 5 mm,4 mm, and 3 mm from the top of the combined ultrasound-thermal image illustrated in Figure6b. Mean temperature increases monotonically with time in all four tissue regions. Atemperature elevation of close to 25°C after 5 seconds of laser irradiation is observed in region1 located below the dermal-fat junction (i.e., in skin). No external cooling techniques wereemployed on the porcine tissue specimen. The temperature increases more than 30°C in region2—located above the dermal-fat junction and consisting primarily of fat. At deeper depths infatty tissue (regions 3 and 4), progressively lower temperature increase is observed.

In a separate specimen, the laser heating performed produced a significantly lower temperaturerise (Fig. 7a) compared to Figure 6a. The overlaid temperature map (Fig. 7b) indicates that thetemperature increased by less than 15°C at the end of laser irradiation, in both the dermal andfatty tissue regions.

Thermal Damage AssessmentHistological assessment performed on the specimen shown in Figure 6 illustrated defects inthe subcutaneous adipose tissue marked with various degrees of compression, disruption andfragmentation (Fig. 8a). Distinctive thermal damage with hyalinization, swelling and loss ofbirefringence in the dermal collagen in a wedge-shaped region at the surface was observed.The epithelial cells of the deeply placed glandular ducts are shrunken and hyperchromatic whilethe surrounding fibroadipose tissue was torn and fragmented (Fig. 6b). Histological assessmentwas also performed on the specimen shown in Figure 7. This sample showed negligible thermaldamage to the adipose tissue layer (Fig. 8c) possibly due to lower temperature rise achievedafter laser irradiation.

DISCUSSIONSubcutaneous fat can be targeted for laser therapy by selecting a wavelength where theabsorption of fat exceeds that of water [7]. However prior to performing laser therapy for fatreduction, identifying the laser dosimetry is important. Our results indicate that ultrasoundimaging in combination with laser irradiation may be utilized to identify the dermis-fat junctionand thereby differentiate between water-based and lipid-based tissues (Fig. 4 and Fig. 5).

To identify the tissue composition, a small temperature increase in response to laser irradiationis needed. Since a single fiber was used to deliver the radiant energy (Fig. 2), the dermis-fatjunction was identified in a relatively small (<5 mm) region. Using a multi-fiber deliverysystem or performing sequential scanning with sub-therapeutic laser dose, the entire region ofinterest can be safely interrogated and a complete tissue composition map generated.

To ensure irreversible thermal damage, the temperature in the therapeutic zone has to bemaintained greater than 43°C for an extended period of time [10,11]. The temperature increasein the subcutaneous fat due to laser irradiation depends on several factors including tissueoptical properties and dermis thickness. For example, for the same irradiation parameters,tissue samples from separate animals had varying temperature profiles at the end of laserheating (Fig. 6 and Fig. 7) leading to different thermal damage outcomes (Fig. 8). Therefore,it is necessary to monitor the temperature increase during the laser treatment. Ultrasound-basedthermal images (Fig. 6 and Fig. 7) indicate the feasibility of performing spatial and temporalmapping of temperature increase during laser irradiation.

In the preliminary experiments performed in this study, a significant temperature increase wasobtained in the dermal region (Fig. 6c). Temperature increase in the dermal region was lowerthan the fat region, possibly due to lipid having higher optical absorption coefficient as

Shah et al. Page 6


NIH


NIH


NIH


compared to water at 1,210 nm laser irradiation wave-length. In a clinical application, periodicsurface cryogen spray cooling can be employed to protect the dermis from thermal damage[26]. In addition, ultrasound thermal imaging may also be utilized to monitor the temperaturein dermis and sub-dermal regions in response to laser irradiation and possibly trigger theperiodic bursts of cooling spray.

Histological evaluation of the samples showed thermal damage at the epidermal and subjacentdermal layers in one specimen (Fig. 8a–b). Another specimen contained subcutaneous defectsthat could be formed by thermal desiccation or various other mechanisms (Fig. 8c).Subcutaneous tears could be due to thermal damage and tissue desiccation possiblycomplicated by incomplete fixation, paraffin penetration and/or sectioning artifacts. However,distinguishing between the two mechanisms is difficult since they produce similar defects[27]. Further studies are needed to identify the laser dosimetry required to ensure thermaldamage.

In experiments reported here, laser irradiation and the ultrasound transducer were on theopposite sides of the tissue sample (Fig. 2a). This geometery led to reverberations on the bottomof the ultrasound image (Fig. 3a), introducing artifacts making tissue identification from theholder more difficult. For in vivo studies light delivery and ultrasound transducer must be onthe same side. Optical fibers placed alongside the transducer can be used for delivering theradiant energy to the tissue, these setups have been assembled for photoacoustic imaging[28,29]. Another alternative is to integrate the ultrasonic transducer and the optical probe intoone assembly similar to confocally arranged transducers used during high intensity focusedultrasound treatments [30].

Physiological motion such as cardiac or respiratory movement could lead to artifacts in theultrasound measurements if it produces normalized time shift in ultrasound images. Forexample, periodic heart beat causes tissue motion which could appear as time shifts in theultrasound signals leading to errors in tissue composition and temperature maps. To avoid theimpact of cardiac-induced motion, it may be necessary to utilize an electrocardiogram (ECG)to trigger the data capture. Ultrasound frames will then be collected at the same point in thecardiac cycle and potentially minimizing tissue-motion artifacts [31].

For remote temperature assessment, a water-bath experiment was first performed to establishthe relationship between normalized time shift and temperature (Fig. 3) for the porcine tissuesample. In ultrasound imaging temperature measurement is possible using a generalized andknown a priori tissue specific calibration [20]. Identifying the tissue composition willpotentially allow the calculation of temperature from ultrasound time shifts directly, accuratelyand in real time without a calibration procedure.

SUMMARYThe results of our study outline the ability of ultrasound imaging to guide and monitor lasertherapy of fat. Ultrasound imaging was used to identity the dermis-fat boundary in porcinetissue with high contrast and to compute the temperature elevations during laser heating.Application of the ultrasound technique reported here may be relevant to clinical laserprocedures to reduce fat.

ACKNOWLEDGMENTSSupport in part by the National Institutes of Health under grants EB008101 and EB004963 is gratefully acknowledged.The authors also would like to thank Mrs. Suhyun Park and Mr. Jared Mendeloff of the Ultrasound Imaging andTherapeutics Research Laboratory at the University of Texas at Austin for their help with the data capture, theultrasound beamforming and image reconstruction.

Shah et al. Page 7


NIH


NIH


NIH


REFERENCES1. Heymans O, Castus P, Grandjean FX, Van Zele D. Liposuction: Review of the techniques, innovations

and applications. Acta Chir Belg 2006;106(6):647–653. [PubMed: 17290687]2. Neira R, Arroyave J, Ramirez H, Ortiz CL, Solarte E, Sequeda F, Gutierrez MI. Fat liquefaction: Effect

of low-level laser energy on adipose tissue. Plast Reconstr Surg 2002;110(3):912–922. [PubMed:12172159]discussion 923–915.

3. Alberto G. Submental Nd:Yag laser-assisted liposuction. Lasers Surg Med 2006;38(3):181–184.[PubMed: 16453321]

4. Katz B, McBean J. The new laser liposuction for men. Dermatol Ther 2008;20(6):448–451. [PubMed:18093018]

5. Toledo LS, Mauad R. Complications of body sculpture: Prevention and treatment. Clin Plast Surg2006;33(1):1–11. v. [PubMed: 16427969]

6. Lehnhardt M, Homann HH, Daigeler A, Hauser J, Palka P, Steinau HU. Major and lethal complicationsof liposuction: A review of 72 cases in Germany between 1998 and 2002. Plast Reconstr Surg 2008;121(6):396e–403e.

7. Anderson RR, Farinelli W, Laubach H, Manstein D, Yaroslavsky AN, Gubeli JG III, Jordan K, NeilGR, Shinn M, Chandler W, Williams GP, Benson SV, Douglas DR, Dylla HF. Selectivephotothermolysis of lipid-rich tissues: A free electron laser study. Lasers Surg Med 2006;38(10):913–919. [PubMed: 17163478]

8. Altshuler GB, Anderson RR, Manstein D, Zenzie HH, Smirnov MZ. Extended theory of selectivephotothermolysis. Lasers Surg Med 2001;29(5):416–432. [PubMed: 11891730]

9. Illouz YG. Study of subcutaneous fat. Aesthetic Plast Surg 1990;14(1):165–177. [PubMed: 2205085]10. Thomsen S. Pathologic analysis of photothermal and photomechanical effects of laser-tissue

interactions. Photochem Photobiol 1991;53(6):825–835. [PubMed: 1886941]11. Badin AZED, Gondek LBE, Garcia MJ, Valle LCD, Flizikowski FBZ, Noronha LD. Analysis of laser

lipolysis effects on human tissue samples obtained from liposuction. Aesthetic Plastic Surgery2005;29(4):281–286. [PubMed: 16075355]

12. Karlan BY, Platt LD. Ovarian cancer screening. The role of ultrasound in early detection. Cancer1995;76:2011–2015. [PubMed: 8634993]

13. Teh W, Wilson ARM. The role of ultrasound in breast cancer screening. A consensus statement bythe European Group for breast cancer screening. Eur J Cancer 1998;34(4):449–450. [PubMed:9713292]

14. Pereira FR, Machado JC, Foster FS. Ultrasound characterization of coronary artery wall in vitro usingtemperature-dependent wave speed. Ultrasonics, Ferroelectrics and Frequency Control. IEEETransactions on 2003;50(11):1474–1485.

15. Shi Y, Witte RS, Milas SM, Neiss JH, Chen XC, Cain CA, O’Donnell M. Ultrasonic thermal imagingof microwave absorption. Proceeding of the 2003 IEEE Ultrasonics Symposium 2003;1:224–227.

16. Shah J, Aglyamov SR, Sokolov K, Milner TE, Emelianov SY. Ultrasound-based thermal and elasticityimaging to assist photothermal cancer therapy—Preliminary study. Proceeding of the 2006 IEEEUltrasonics Symposium 2006:1029–1032.

17. Shah J, Park S, Aglyamov S, Larson T, Ma L, Sokolov K, Johnston K, Milner T, Emelianov S.Photoacoustic and ultrasound imaging to guide photothermal therapy: Ex-vivo study. Proc SPIE2008;6856:68560U.

18. Shah J, Aglyamov SR, Sokolov K, Milner TE, Emelianov SY. Ultrasound imaging to monitorphotothermal therapy— Feasibility study. Opt Express 2008;16(6):3776–3785. [PubMed: 18542473]

19. Seip R, Ebbini ES. Non-invasive monitoring of ultrasound phased array hyperthermia and surgerytreatments. Proceedings of IEEE 17th Annual conference on Engineering in Medicine and BiologySociety 1995;661:663–664.

20. Varghese T, Zagzebski JA, Chen Q, Techavipoo U, Frank G, Johnson C, Wright A, Lee FT Jr.Ultrasound monitoring of temperature change during radiofrequency ablation: Preliminary in-vivoresults. Ultrasound Med Biol 2002;28(3):321–329. [PubMed: 11978412]

21. Maass-Moreno R, Damianou CA. Noninvasive temperature estimation in tissue via ultrasound echo-shifts. Part I. Analytical model. J Acoust Soc Am 1996;100(4 Pt 1):2514–2521. [PubMed: 8865654]

Shah et al. Page 8


NIH


NIH


NIH


22. Maass-Moreno R, Damianou CA, Sanghvi NT. Noninvasive temperature estimation in tissue viaultrasound echo-shifts. Part II. In vitro study. J Acoust Soc Am 1996;100(4 Pt 1):2522–2530.[PubMed: 8865655]

23. Duck, FA. Physical properties of tissue. New York: Academic; 1990.24. Bamber JC, Hill CR. Ultrasonic attenuation and propagation speed in mammalian tissues as a function

of temperature. Ultrasound Med Biol 1979;5:149–157. [PubMed: 505616]25. Lubinski MA, Emelianov SY, O’Donnell M. Speckle tracking methods for ultrasonic elasticity

imaging using short-time correlation. Ultrasonics, Ferroelectrics and Frequency Control. IEEETransactions on 1999;46(1):82–96.

26. Anvari B, Tanenbaum BS, Milner TE, Tang K, Liaw L-H, Kalafus K, Kimel S, Nelson JS. Spatiallyselective photocoagulation of biological tissues: Feasibility study utilizing cryogen spray cooling.Appl Opt 1996;35(19):3314–3320.

27. Thomsen SL. Practical pathology for engineers: How to do the job right the first time. Proc SPIE2003;4954(1):476635.

28. Zemp RJ, Bitton R, Li M-L, Shung KK, Stoica G, Wang LV. Photoacoustic imaging of themicrovasculature with a high-frequency ultrasound array transducer. J Biomed Opt 2007;12(1):010501. [PubMed: 17343475]

29. Park S, Mallidi S, Karpiouk AB, Aglyamov S, Emelianov SY. Photoacoustic imaging using arraytransducer. Proc SPIE 2007;6437:643714.

30. Seip R, Ebbini ES. Noninvasive estimation of tissue temperature response to heating fields usingdiagnostic ultrasound. Biomed Eng IEEE Trans 1995;42(8):828–839.

31. Simon C, VanBaren PD, Ebbini ES. Motion compensation algorithm for noninvasive two-dimensional temperature estimation using diagnostic pulse-echo ultrasound. SPIE Surg Appl Energy1998;3249:182–192.

Shah et al. Page 9


NIH


NIH


NIH


Fig. 1.Block diagram illustrating the principles of ultrasound measurements for identifying tissuecomposition and for thermal imaging.

Shah et al. Page 10


NIH


NIH


NIH


Fig. 2.a: Experimental setup for ultrasound imaging during laser heating. b: Digital photograph ofthe experimental setup showing the orientation of the laser fiber, ex vivo tissue and ultrasoundtransducer. c: Block diagram for computing grayscale B-mode ultrasound image, tissuecomposition map and thermal image of the tissue sample.

Shah et al. Page 11


NIH


NIH


NIH


Fig. 3.Temperature calibration for (a) porcine fat and (b) porcine skin. Note, the negative temperaturegradient of the normalized time shift for fatty tissue and positive temperature gradient for water-based tissue.

Shah et al. Page 12


NIH


NIH


NIH


Fig. 4.a: Ultrasound image of the porcine tissue. Image covers a 10 mm (depth) by 15 mm (width)region. b: Normalized time shift between the arrows in Figure 3a after 1, 3, and 5 seconds oflaser heating. The zero crossing indicates the dermis-fat junction.

Shah et al. Page 13


NIH


NIH


NIH


Fig. 5.a: Map of the normalized time shift after 5 seconds of laser irradiation with clear demarcationbetween positive and negative normalized time shift under laser irradiation region. b:Ultrasound image of the porcine tissue with the zero-crossing of normalized time shiftsuperposed to represent the dermis-fat junction. All images represent a 10 mm by 15 mm region.

Shah et al. Page 14


NIH


NIH


NIH


Fig. 6.a: Thermal image showing the temperature elevation reached due to laser exposure. b:Ultrasound image overlaid with the temperature maps, showing the temperature elevation inthe dermal and fatty regions. c: Temporal temperature rise at four regions directly along thelaser irradiation plane. The regions are shown as boxes in (b).

Shah et al. Page 15


NIH


NIH


NIH


Fig. 7.a: Thermal image showing the temperature elevation reached due to laser exposure. b:Ultrasound image overlaid with the temperature maps, showing the temperature elevation inthe dermal and fatty regions.

Shah et al. Page 16


NIH


NIH


NIH


Fig. 8.a: Porcine skin overview showing defects in the subcutaneous adipose tissue (thick arrows).Wedge shaped surface lesion (thin arrows) visible showing signs of thermal denaturation ofcellular structural proteins [H&E stains. Orig. Mag. 16×]. b: The glandular ducts showcompression and are hyper chromatic. The adipose tissue (arrows) is torn and fragmentedassociated with subcutaneous defects. c: Normal glandular ducts surrounded by compressedfat cells (arrows) in a specimen with lower temperature increase (less than 15°C) [H&E stains.Orig. Mag. 200×].

Shah et al. Page 17


NIH


NIH


NIH


Nihms116835

Business

laser irradiation

fat damage

ultrasound imaging system

monitoring of laser

theory ultrasound imaging

porcinesubcutaneous

fat breakdown

fat treatments