PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
IEEE Trans Ultrason Ferroelectr Freq Control. Author manuscript; available in PMC 2017 August 10.
Published in final edited form as:
PMCID: PMC5551400
NIHMSID: NIHMS884545

Quantitative Ultrasound for Monitoring High-Intensity Focused Ultrasound Treatment In Vivo

Abstract

The success of any minimally invasive treatment procedure can be enhanced significantly if combined with a robust noninvasive imaging modality that can monitor therapy in real time. Quantitative ultrasound (QUS) imaging has been widely investigated for monitoring various treatment responses such as chemotherapy, radiation, and thermal therapy. Previously, we demonstrated the feasibility of using spectral-based QUS parameters to monitor high-intensity focused ultrasound (HIFU) treatment of in situ tumors in euthanized rats [Ultrasonic Imaging 36(4), 239–255, 2014]. In the present study, we examined the use of spectral-based QUS parameters to monitor HIFU treatment of in vivo rat mammary adenocarcinoma tumors (MAT) where significant tissue motion was present. HIFU was applied to tumors in rats using a single-element transducer. During the off part of the HIFU duty cycle, ultrasound backscatter was recorded from the tumors using a linear array co-aligned with the HIFU focus. A total of 10 rats were treated with HIFU in this study with an additional sham-treated rat. Spectral parameters from the backscatter coefficient, i.e., effective scatterer diameter (ESD) and effective acoustic concentration (EAC), were estimated. The changes of each parameter during treatment were compared with a temperature profile recorded by a fine-needle thermocouple inserted into the tumor a few millimeters behind the focus of the HIFU transducer. The mean ESD changed from 121 ± 6 to 81 ± 8 μm(p-value = 0.0002), and the EAC changed from 33 ± 2 to 46 ± 3 dB/cm3 (p-value = 0.0002) during HIFU exposure as the temperature increased on average from 38.7 ± 1.0° C to 64.2 ± 2.7° C. The changes in ESD and EAC were linearly correlated with the changes in tissue temperature during the treatment. When HIFU was turned off, the ESD increased from 81 ± 8 to 121 ± 7 μm and the EAC dropped from 46 ± 3 to 36 ± 2 dB/cm3 as the temperature decreased from 64.2 ± 2.7° C to 45 ± 2.7° C. QUS was demonstrated in vivo to track temperature elevations caused by HIFU exposure.

Index Terms: Backscatter coefficient (BSC), focused ultrasound, quantitative ultrasound (QUS) imaging, ultrasound-based therapy

I. Introduction

High-intensity focused ultrasound (HIFU) is a noninvasive thermal therapy approach for treating tissue for various diseases including cancer. HIFU treatment can heat or ablate tissue in the focal region of the transducer without damaging the surrounding healthy tissue. Thermal therapy remains an important cancer therapy option because it not only kills the cancer cells but can also activate anticancer immunity in the body [1].

HIFU is a noninvasive procedure with various advantages compared to invasive procedures, such as less blood loss, decreased pain and need for postoperative medications, quick patient recovery, and reduced scarring. HIFU is currently being tested and, in some cases, adopted for clinical therapy [2]. Several investigators have reviewed clinical studies involving HIFU to treat diseases [2], [3]. Currently, HIFU therapy has been approved in the United States for the treatment of uterine fibroids and bone metastases [4], [5]. HIFU is being investigated for treatment of various diseases such as the cancer of the prostate, breast, liver, and to treat Parkinson’s disease.

The recent success of HIFU treatment largely depends on the availability of a robust imaging and monitoring system [6]. Various imaging modalities such as ultrasound, X-ray CT, and MRI have been investigated for guiding and monitoring HIFU treatment. Currently, magnetic resonance thermal imaging (MRTI) is considered as the best-suited imaging modality for guiding HIFU (MRgHIFU). An accuracy of 1° C and update time of 1 s has been reported for MRI [7]. The spatial resolution of MRI thermometry imaging is around 2 mm in stationary tissue. Proton-resonance frequency approaches to MRI have achieved near-real time (i.e., 10 Hz) imaging [8]–[10]. However, these techniques require predictable motion patterns in order to register image frames. Finally, MRI is expensive and requires an MRI compatible HIFU system for the treatment. Moreover, MRI cannot currently provide a real-time monitoring capability and may inaccurately estimate tissue temperature with motion artifacts or low signal-to-noise ratio (SNR). Finally, the accessibility of MRgHIFU is limited in North America.

Significant research has been conducted to develop robust, portable, accessible, and low-cost imaging methods for monitoring HIFU treatments [11]–[14]. Foremost among these are ultrasound imaging techniques, which have been widely investigated for monitoring ablation techniques including HIFU treatments. Ultrasound imaging is low cost, uses nonionizing radiation, has real-time capabilities, and is portable with the potential to increase accessibility [15]. Quantitative ultrasound (QUS) imaging techniques involving estimation of changes in sound speed, backscatter, attenuation, stiffness, and strain in tissue have been investigated to monitor thermal therapies. Changes in sound speed have been correlated to temperature changes during HIFU exposure [16], [17]. Some investigators have explored spectral-based ultrasound techniques to monitor the temperature changes in tissues treated with HIFU [18]–[20]. Changes in the backscatter coefficient (BSC) have been observed in coagulated tissue formed by HIFU exposure when compared with normal tissue [21]. Ultrasound attenuation and backscatter have been used to detect lesion formation during HIFU exposure [22]. However, all of these techniques have yet to be adopted clinically and issues with tissue motion or nonlinear changes in the parameter space versus temperature may limit their adoptability. However, it must be noted that recent advances in adaptive filter design for motion compensation and high frame rate imaging have alleviated some of the sensitivity to tissue motion of frame-to-frame correlation-based methods for estimating temperature in tissues [23]–[25].

As an alternative, we have been developing QUS imaging techniques that can be used for both assessment [26]–[28] and monitoring of HIFU therapy [29]–[31]. In experiments using biological phantoms and fresh liver samples in a saline bath, changes in spectral-based QUS parameters [i.e., the effective scatterer diameter (ESD) and effective acoustic concentration (EAC)] were correlated to temperature elevations in tissues [30], [32]. The accumulation of research findings by several groups clearly indicates that heating of tissues, such as that occurs with HIFU exposure, induces changes in tissue microstructure. Because ultrasound scattering depends on the properties of the propagating media, HIFU can be monitored by analyzing the backscattered signal during exposure. In our previous study, we examined spectral-based QUS parameters for their ability to track temperature elevations in tumors in euthanized rats induced by HIFU with no tissue motion or perfusion [29]. In the present study, we are extending our previous work by using spectral-based QUS parameters (i.e., the ESD and EAC) to monitor in vivo HIFU treatment in a rat mammary adenocarcinoma where significant tissue motion is present. The changes in the various spectral-based QUS parameters were monitored during the HIFU exposure and compared with the temperature profile recorded from a fine needle thermocouple inserted into the tumor behind the focus of the HIFU transducer. The shape and the amplitude of the frequency-dependent BSC were analyzed to monitor in vivo HIFU exposure.

II. Methods

A. Animal Model and Handling

The experimental protocol was approved by the Institutional Animal Care and Use Committee (IACUC), University of Illinois, Urbana-Champaign and satisfied all University and NIH rules for the humane use of laboratory animals. All the animals (N = 11) were provided food ad libitum. Mammary adenocarcinoma tumors (MAT B III) cells [American Type Culture Collection (ATCC), Manassas, VA, USA] were cultured and used for growing mammary tumors in rats. The cells were cultured in ATCC-formulated McCoy’s 5a Medium Modified along with 10% fetal bovine serum (HyClone Laboratories, Logan, UT, USA) and 1% penicillin-streptomycin (HyClone Laboratories, Logan, UT, USA). Approximately, 500 cells were suspended in saline (100 μL) and injected subcutaneously into the fat pad of a rat to grow a tumor. Specifically, female 10–21 week-old Fischer 344 rats (Harlan Laboratories, Inc., Indianapolis, IN, USA) were used in the experiments. At the time of injection of the MAT cells, the weight of the rats was recorded. When a tumor reached a size of 8–9 mm in diameter, the ultrasound experiment was conducted. During a HIFU exposure, the rat was anesthetized using isoflurane gas through a mask covering the mouth and nose of the rat. The hair over the tumor was shaved and depilated to maximize sound transmission. The rats were placed on a specially designed holder and partially submerged in a tank of degassed water maintained at 37° C such that the tumor was under water.

B. Ultrasound Method

Ten of the rat tumors were treated with HIFU and one rat tumor was sham treated. The sham-treated rat was used to quantify the effects of breathing on QUS estimates without HIFU. All experiments were conducted in vivo. The HIFU system consisted of a 1-MHz (f/1.1) transducer with aperture diameter of 4.7 cm. An arbitrary waveform generator (HP 33120a, Agilent Technologies, Santa Clara, CA, USA) was used to generate a long duty cycle tone burst. The tone burst was amplified by an A150-55-dB power amplifier (ENI, Rochester, NY, USA). The −6 dB beamwidth and −6 dB depth of field of the HIFU transducer was characterized using a needle hydrophone (Precision Acoustics, Dorchester, U.K.) in degassed water at 37° C. The −6 dB beamwidth and depth of field were 1.8 mm and 5.2 cm, respectively. During the off cycle of the duty cycle, an L14-5/38 imaging probe attached to a SonixRP ultrasound imaging system (Ultrasonix, Richmond, BC, Canada) recorded a single ultrasonic imaging frame. The SonixRP system provided the raw radio frequency (RF) signals at a sampling frequency of 40 MHz. The imaging array had a center frequency of 6 MHz as estimated from recorded signals in pulse-echo mode from a planar reflector. An analysis bandwidth of 4–9 MHz was used in the subsequent spectral analysis.

A custom holder was machined to align the focus of the HIFU transducer with the imaging plane of the imaging array. Once the HIFU transducer and imaging array were placed in the holder, no further adjustment was needed. Fig. 1 illustrates the experimental configuration and the custom holder that was machined. To continuously estimate temperature in the tumor, a needle thermocouple (HYP1-30-1/2-T-G-60-SMPWM, Omega Engineering, Inc., Stamford, CT, USA) was inserted into the tumor behind the focus of the HIFU source. The thermocouple was placed behind the focus of the HIFU source for two reasons: 1) this placed the thermocouple out of the way of the region undergoing RF analysis and 2) to reduce the presence of viscous heating artifacts [33]. The tumor is identified by the red dot in Fig. 1(a). The needle thermocouple had a diameter of 0.3 mm and a length of 15 mm. The output of the thermocouple was connected to a temperature reader (NI USB-TC01, National Instruments Corporation, Austin, TX, USA). The reader was connected to a computer and the temperature was recorded to hard drive once every second. In order to maintain the temperature of the degassed water bath, in which the rat was partially submerged, at 37° C a heating element was placed in the bath and controlled by an automated temperature controller. A computer-controlled micropositioning system (Parker Hannifin Corporation, Rohnert Park, CA, USA) was used during the experiment to move the HIFU transducer and imaging probe.

Fig. 1
(a) Experimental configuration. (b) Custom holder made for the HIFU and imaging array. (c) Illustration of HIFU and imaging sequences.

HIFU was used to elevate the temperature in the tumor. A spatial peak temporal average intensity (ISPTA) of 502 W/cm2 was used in the HIFU exposures with 75% duty cycle and 30 Hz pulse repetition frequency. The intensity of 502 W/cm2 was used based on our previous studies, where visible tissue thermal damage was observed from histopathology analysis [43]. The ISPTA was estimated from pressure measurements at the focus of the HIFU source in degassed water using a needle hydrophone (Precision Acoustics, Dorchester, U.K.). RF imaging frames of the tumor were collected by the SonixRP system during the therapy by synchronizing the acquisition of RF imaging frames with the off portion of the HIFU exposure duty cycle [see Fig. 1(c)]. At the end of each HIFU tone burst, the SonixRP was triggered on to record a single RF imaging frame. The acquisition of the RF data was required to calculate the BSC from the tumor. The tumors were exposed for a duration of 60 s.

C. Processing Procedures

B-mode images of a tumor intersecting the HIFU focus were constructed from the RF data acquired by the SonixRP. A region of interest (ROI) was manually drawn in the B-mode image corresponding to the tumor area. The ROI was used to localize the tumor region for the spectral analysis of the backscattered RF echoes. The pulse length (PL) was measured as 0.37 mm. Data blocks of size 7 PL × 7 PL were constructed within the ROI. The data blocks in the ROI had an 80% overlap between adjacent data blocks both axially and laterally. Bias and variance of QUS estimates are related to the time-bandwidth product of the range gate length (axial data block size) [34]. Therefore, the data block size was quoted in terms of the number of PLs. A data block 7 pulse lengths × 7 pulse lengths was chosen to provide good QUS estimates in terms of low bias and low variance versus data block size. An 80% overlap of data blocks provides the ability to enlarge the data blocks while reducing the pixel sizes of the QUS image maps.

Fig. 2 shows an image depicting the selection of the data blocks within the ROI. First, the ROI, corresponding to the tumor boundary, was outlined as depicted by the red line in Fig. 2. Next, data blocks were created inside the ROI boundary using a custom MATLAB (MathWorks, Natick, MA, USA) code to automate the selection of the data blocks. The BSC was estimated for each scan line in a data block and averaged to provide the BSC estimate for the data block. Calibration of the BSC was accomplished using the reference phantom technique and a well-characterized phantom [35]. Specifically, the BSC was calculated from the averaged power spectrum from the windowed scan lines from a data block. This averaged power spectrum was divided by an averaged power spectrum calculated from a data block at the same depth in the well-characterized reference phantom. To estimate the BSC, the averaged power spectrum from the sample data block was divided by the averaged spectrum from the reference phantom data block and corrected for attenuation differences between the sample and phantom. ESD and EAC were then estimated from the BSC for each data block using a spherical Gaussian model [36].

Fig. 2
Illustration of data block selection from a tumor (tumor outlined in red).

In a separate set of experiments, the insertion loss technique was used to estimate attenuation from untreated excised tumors (n = 4) [37]. In these experiments, the tumor from a rat was excised and any overlaying skin removed prior to scanning. An attenuation slope was estimated for each of the excised tumor samples and averaged. The average attenuation slope of 0.7 ± 0.1 dB/cm/MHz was used to compensate the power spectra for losses in the RF signals due to frequency-dependent attenuation.

The average attenuation slope was used to compensate the power spectrum for frequency-dependent losses for each recorded RF frame in the in vivo HIFU experiments. Because we used a single attenuation value throughout the in vivo experiments, differences in the attenuation due to biological variability between rat tumors, temperature elevations, or from irreversible tissue changes were not used in the compensation of the BSC. However, the size of the tumors was less than 1 cm; therefore, differences in attenuation from the average attenuation would not result in large errors in BSC estimates over the frequency ranges used in the analysis. Based on results from our previous in situ study, it was assumed that attenuation effects would be small for the current study [29].

A spherical Gaussian scattering model and rapid estimator was used to extract ESD and EAC estimates from the BSC for each data block in a tumor [36], [38]. The change in ESD and EAC for each data block was estimated with respect to its initial value before the HIFU exposure. For example, the change in EAC after 5 s is given by

ΔEAC=EAC(5s)-EAC(0s)EAC(0s)
(1)

where EAC (0 s)and EAC (5 s) refer to EAC at the beginning of the exposure just before HIFU exposure and 5 s after the start of the HIFU exposure, respectively. The changes in the ESD estimates versus exposure time were similarly calculated.

The changes in ESD and EAC were assessed through a Kruskal–Wallis one-way analysis of variance by ranks when comparing the estimates at three different time points. Statistically significant differences were quantified through p-values. Case I corresponded to a time point just before exposure. Case II corresponded to a time point just before the last HIFU tone burst was delivered (i.e., the peak temperature). Finally, Case III corresponded to a time point after the tissue was cooled to post exposure. Based on continuous thermocouple readings from the tumor, a return to 37° C typically occurred 5–10 min after the HIFU was turned off. In addition, linear curve fitting was performed between the recorded temperature (x-variable) and ΔESD or ΔEAC (y-variable). Specifically, the relationship between the ESD and EAC parameters and temperature during the exposure was quantified both for the heating period (HIFU on) and during the cooling period (HIFU off).

III. Results

Fig. 3 shows representative B-mode images and QUS images enhanced by ESD and EAC of a rat tumor at different time points during HIFU exposure. The temperature profile recorded by the thermocouple and the changes in the ESD and EAC as a function of time are shown in Fig. 3 (right). The ESD and EAC curves were generated by averaging the QUS parameters over the whole ROI and calculating the percent difference between the QUS parameter at each time point and the QUS parameter average at baseline, i.e., before HIFU was turned on. Examination of the images reveals that significant motion occurred during the HIFU exposure due to breathing. Because the rat is a small animal and the tumor was on the abdominal wall, large motion out of plane both laterally and vertically was observed. The jagged nature of the QUS parameters as the temperature increased is hypothesized to be due in part to the motion of the tumor during exposure. In all animals, the EAC was observed to increase with temperature and decrease once the temperature began to decrease.

Fig. 3
(Top) B-mode images at different time points. (Middle) Series of ΔESD images of tumor during treatment with HIFU and corresponding temperature/ΔESD curves versus time. (Bottom) Series of ΔEAC images of tumor during treatment with ...

The changes in QUS parameters due to the breathing cycle of the rat alone were quantified using the same experimental setup shown in Fig. 1 with no HIFU as a sham experiment. Changes in ESD and EAC were less than 5% for a period of 120 s as shown in Fig. 4. In the sham exposure, no trends in the QUS parameters were observed; however, small fluctuations were observed in the QUS parameters during the sham exposure. These variations are mainly due to motion of the tumor during imaging and would partially account for some of the observed fluctuations observed in the QUS images created during and after HIFU exposure.

Fig. 4
Changes in (a) ESD and (b) EAC versus time from rat tumor with no HIFU.

For all the animals, the ESD and EAC parameters were correlated to temperature changes during the HIFU treatment. The increases in temperature during the exposure and the decreases in temperature during the cool down period were correlated to the changes in ESD and EAC parameters. A linear fit between changes in QUS parameters (ΔESD and ΔEAC) and temperature recorded by the thermocouple for the heating and cooling period for each experiment were performed. Example linear fits between the QUS parameters and tissue temperature during heating and cooling periods are shown in Figs. 5 and and66 for two tumors (labeled R1 and R2). The ESD decreased with increasing temperature and increased with decreasing temperature, indicating an inverse relationship with changes in tissue temperature. On the other hand, the EAC increased with increasing temperature and decreased with decreasing temperature. The slope, intercept, and the R2 of the linear fit are tabulated in Table I for all the tumors used in the current study. The slopes of the linear fit between the temperature and the ESD during the heating and cooling periods were −1.22 ± 0.16 and −1.21 ± 0.26, respectively. The slopes of the linear fit between the temperature and EAC during the heating and cooling periods were 1.48 ± 0.26 and 1.51 ± 0.27, respectively. The goodness-of-fit between the changes in ESD and temperature was measured using R2 values which were 0.90 ± 0.05 and 0.75 ± 0.05 for the heating and cooling periods, respectively. Similarly, the R2 between changes in EAC and temperature were 0.92 ± 0.04 and 0.85 ± 0.03 for the heating and cooling periods, respectively.

Fig. 5
Best fit line of temperature versus ΔESD for (a) R1 during heating; (b) R2 during heating; (c) R1 during cooling; and (d) R2 during cooling phase of the HIFU experiment.
Fig. 6
Best fit line of temperature versus ΔEAC for (a) R1 during heating; (b) R2 during heating; (c) R1 during cooling; and (d) R2 during cooling phase of the HIFU experiment.
TABLE I
Parameters of the Linear Fits of Temperature Recorded by Thermocouple Versus ΔESD (%) and ΔEAC (%)

Statistical analysis was conducted to investigate the significance of the estimated ESD and EAC at different time periods during the HIFU treatment (Cases I, II, and III). The mean ESD and EAC values before HIFU exposure (Case I) were 121 ± 6 μm and 33 ± 2 dB/cm3, respectively. Just before the last HIFU tone burst was delivered to the tissue (Case II), the ESD and EAC values were 81 ± 8 μm and 46 ± 3 dB/cm3, respectively. After the tissue cooled down to body temperature of approximately 37° C as recorded by the thermocouple (Case III), the ESD and EAC values were 120 ± 7 μm and 36 ± 2 dB/cm3, respectively. Statistically significant differences were observed between Case I and Case II in the average ESD (p-value = 0.0002) and EAC (p-value = 0.0002) estimates. No statistically significant differences were observed between Cases I and III (p-value = 0.5956 and p-value = 0.0833 for ESD and EAC, respectively). The changes in the ESD and EAC parameters significantly changed with increases in tissue temperature during HIFU treatment and then returned to pretreatment values after cooling off.

IV. Discussion

QUS imaging was used to monitor HIFU treatment of rat tumors in vivo. A needle thermocouple was positioned in the tumor and behind the focus of the HIFU source to measure temperature in the tumor tissue. An ultrasound array and SonixRP imaging system were used to guide the placement of the needle thermocouple and also to acquire raw RF signals for QUS imaging. The changes in the QUS parameters were correlated with the changes in the temperature during the HIFU exposures. Because the thermocouple was placed behind the focus and at the back of the tumor, it was not possible to associate specific values of ESD or EAC with a specific temperature. More than likely, higher temperature elevations were occurring in the actual regions being analyzed with QUS than were recorded by the thermocouple.

The results demonstrated good correlation between the spectral-based QUS parameters and the changes in tissue temperature during the HIFU exposure and also during the cool-down period after the HIFU was turned off. It was observed that tissue motion due to the breathing cycle had small but noticeable influences on the changes in the spectral-based QUS parameters. The results from the sham-exposure experiments exhibited small changes of around 5% in spectral-based QUS parameters as time elapsed suggesting that the QUS imaging technique is robust against tissue motion effects. These small fluctuations in the sham exposure were comparable in size to fluctuations observed in the in vivo HIFU-exposure experiments. Similar fluctuations were not observed in the in situ experiments in dead rats [29]. Small fluctuations were observed in the results due to motion, which can be further minimized by averaging RF signals from several frames, i.e., averaging over slow time. Averaging of frames in slow time could further alleviate the artifacts observed in QUS thermometry over time. This would provide a tradeoff between reduced estimate variance and frame rate.

While the QUS parameters were affected by tissue motion, the underlying trends, i.e., that the QUS parameters changed with temperature, were still observable. The estimates of the spectral-based QUS parameters do not depend on the previous frame of ultrasound to deduce their properties like techniques to estimate temperature elevations from changes in sound speed [18]. The scattering properties (i.e., ESD and EAC) of the tissue microstructure change during temperature elevation and these are absolute properties of the underlying medium, which can be deduced from analysis of the RF-backscattered signals from any individual frame.

The changes in the ESD and EAC provided good correlation with changes in tissue temperature both during the heating period while the HIFU was turned on and during the cooling period after the HIFU was turned off as observed quantitatively from the slopes of the linear fit between the spectral-based QUS parameters and the changes in temperature. The R2 values shown in Table I suggest that the spectral-based QUS parameters closely tracked the changes in temperature recorded by the thermocouple. A linear fit was initially chosen due to the lack of availability of a better model. However, during the heating period, the R2 values for all rats were above 0.81 and most of these were in the 0.90 and above range. This suggests that assuming a linear fit allowed 81% or more of the observed changes in ESD and EAC to be explained by the temperature changes. Typically, at the end of the cool-down period, the tissue temperature decreased to the normal body temperature.

While we hypothesize that the changes in the spectral-based QUS parameters was due mainly to the changes in temperature, we cannot rule out mechanical effects from HIFU on the spectral-based QUS estimates versus time. Mechanical effects due to boiling or cavitation could also contribute to the changes in ESD and EAC observed. Additional study will need to be conducted to quantify the contributions of mechanical effects to the observed changes in the spectral-based QUS parameters.

In the tumors, the EAC increased with increasing temperature and decreased with decreasing temperature. In experiments on tumors in dead rats, the EAC was observed to increase with increasing temperature, while the ESD trend with increasing temperature was less clear [29]. The reasons for this lack of a trend in the ESD from tumors in dead rats are presently not understood. However, it is worth reporting that studies using HIFU to treat liver samples have also exhibited a correlation between these parameters and tissue temperature, but with some differences [39]. In liver, the EAC decreased with increasing temperature from HIFU and then increased with decreasing temperature. Similarly, using QUS to monitor microwave ablation of liver samples demonstrated similar behavior for EAC as was observed for HIFU therapy of liver samples [40]. Therefore, the changes observed in the spectral-based QUS parameters versus temperature elevations appear to be tissue-dependent. Understanding the mechanisms behind this tissue-dependent observation is an ongoing investigation.

Finally, previous studies have suggested that the BSC and, therefore, ESD and EAC estimates are relatively insensitive to irreversible changes in tissue properties brought on solely by thermal effects (i.e., no mechanical effects) [41]–[43]. This is promising in the respect that as the tissues undergo thermal coagulation, the BSC changes are still due to changes in temperature, unlike sound speed and attenuation which change their trends versus temperature after thermal coagulation. On the other hand, using the BSC to detect acute thermal coagulation does not appear to be promising.

V. Conclusion

QUS techniques were studied for their potential to track temperature in rat MAT tumors in vivo during HIFU treatment. The EAC parameter increased with increasing temperature and decreased with decreasing temperature in all rat experiments. The ESD was observed to decrease with increasing temperature and then increase with decreasing temperature. The results suggest that spectral-based QUS parameters can monitor temperature elevations due to HIFU therapy in vivo and, furthermore, can be robust against tissue motion.

Acknowledgments

This work was supported by NIH Grant R01 EB008992.

The authors would like to thank the assistance of S. Swat and R. Abuhabshah.

Biographies

An external file that holds a picture, illustration, etc.
Object name is nihms884545b1.gif

Goutam Ghoshal received the B.E. degree in mechanical engineering from the University of Pune, Pune, India, in 2000, and the M.S. and Ph.D. degrees in engineering mechanics from the University of Nebraska-Lincoln, Lincoln, NE, USA, in 2003 and 2008, respectively.

He was a Postdoctoral Research Associate with the University of Illinois at Urbana-Champaign, Champaign, IL, USA, from 2009 to 2012, and a Senior Research Engineer with Acoustic MedSystems Inc., Savoy, IL, USA, conducting research in medical ultrasound. Currently, he is a Senior Ultrasonic R&D Engineer with FloDesign Sonics Inc., Wilbraham, MA, USA, conducting research in ultrasonic application for cell processing and cell therapy. His research interests include biomedical imaging, quantitative ultrasound imaging, therapeutic ultrasound, cell therapy, and computational methods.

Dr. Ghoshal is a member of the Acoustical Society of America and Sigma Xi.

An external file that holds a picture, illustration, etc.
Object name is nihms884545b2.gif

Jeremy P. Kemmerer was born in Allentown, PA, USA, in 1981. He received the B.S. degree in engineering science from Pennsylvania State University, State College, PA, USA, in 2004, and the M.S. and Ph.D. degrees in electrical engineering from the University of Illinois at Urbana-Champaign, Urbana, IL, USA, in 2011 and 2014, respectively.

He is currently a Software Engineer with MathWorks, Natick, MA, USA. His research interests include therapeutic ultrasound, ultrasound image and signal processing, computational methods in acoustics, and quantitative ultrasound.

Dr. Kemmerer was an NIH Fellow from 2013 to 2014.

An external file that holds a picture, illustration, etc.
Object name is nihms884545b3.gif

Chandra Karunakaran received the Bachelor’s degree in instrumentation engineering from Anna University, Chennai, India, in 2005, and the Doctoral degree in biomedical engineering from the University of Cincinnati, Cincinnati, OH, USA, in 2012. She is currently pursuing the M.B.A. degree at Southern Illinois University, Carbondale, IL, USA.

She did her postdoctoral work at the University of Illinois, Urbana-Champaign, Urbana, IL, USA. Her research interests include novel applications of ultrasound, image-processing, medical device innovation and commercialization, and entrepreneurship.

An external file that holds a picture, illustration, etc.
Object name is nihms884545b4.gif

Rita J. Miller received the DVM from the University of Wisconsin, Madison, WI, USA, in 1992. She completed a small animal medical/surgical internship at the University of Illinois at Urbana-Champaign, Urbana, IL, USA, in 1993.

She joined the Bioacoustics Research Laboratory, University of Illinois at Urbana-Champaign, in 1998, where she is a Senior Research Specialist of Bioengineering. Her research interests include ultrasound emphasis, assessment of the biological effects of ultrasound on tissue, the early detection and grading of fatty liver disease, and the interaction of contrast agents with ultrasound.

An external file that holds a picture, illustration, etc.
Object name is nihms884545b5.gif

Michael L. Oelze (M’03–SM’09) was born in Hamilton, New Zealand, in 1971. He received the B.S. degree in physics and mathematics from Harding University, Searcy, AR, USA, in 1994, and the Ph.D. degree in physics from the University of Mississippi, Oxford, MS, USA, in 2000.

From 2000 to 2002, he served as a Postdoc with the Department of Electrical and Computer Engineering (ECE), Bioacoustics Research Laboratory, University of Illinois at Urbana-Champaign (UIUC), Urbana, IL, USA. He joined the Faculty of ECE at the UIUC in 2005 and continues to serve as an Associate Professor. His research interests include quantitative ultrasound imaging, ultrasound bioeffects, ultrasound tomography techniques, ultrasound-based therapy, and application of coded excitation to ultrasonic imaging.

Dr. Oelze is currently a Fellow of the AIUM and a member of ASA. From 2002 to 2004, he was an NIH Fellow conducting research in quantitative ultrasound techniques. He also serves as an Associate Editor of the IEEE Transactions on Ultrasonic, Ferroelectrics, and Frequency Control, Ultrasonic Imaging, and the IEEE Transactions on Biomedical Engineering.

Contributor Information

Goutam Ghoshal, Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801 USA. He is now with FloDesign Sonics Inc., Wilbraham, MA 01095, USA.

Jeremy P. Kemmerer, Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801 USA. He is now with MathWorks, Natick, MA 01760, USA.

Chandra Karunakaran, Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801 USA. She is now with Southern Illinois University, Carbondale, IL 62901, USA.

Rita J. Miller, Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801 USA.

Michael L. Oelze, Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801 USA.

References

1. Lee CT, Mace T, Repasky EA. Hypoxia-driven immunosuppression: A new reason to use thermal therapy in the treatment of cancer? Int J Hyperthermia. 2010;26:232–246. [PMC free article] [PubMed]
2. Kennedy JE. High-intensity focused ultrasound in the treatment of solid tumours. Nat Rev Cancer. 2005 Apr;5:321–327. [PubMed]
3. Zhou YF. High intensity focused ultrasound in clinical tumor ablation. World J Clin Oncol. 2011 Jan 10;2:8–27. [PMC free article] [PubMed]
4. Stewart EA, et al. Focused ultrasound treatment of uterine fibroid tumors: Safety and feasibility of a noninvasive thermoablative technique. Amer J Obstetrics Gynecol. 2003 Jul;189:48–54. [PubMed]
5. Tempany CM, Stewart EA, McDannold N, Quade BJ, Jolesz FA, Hynynen K. MR imaging-guided focused ultrasound surgery of uterine leiomyomas: A feasibility study. Radiology. 2003 Mar;226:897–905. [PubMed]
6. Clement GT. Perspectives in clinical uses of high-frequency focused ultrasound. Ultrasonics. 2004;42:1087–1093. [PubMed]
7. Denis de Senneville B, Quesson B, Moonen CT. Magnetic resonance temperature imaging. Int J Hyperthermia. 2005 Sep;21:515–531. [PubMed]
8. Ries M, et al. Three dimensional motion compensation for real-time MRI guided focused ultrasound treatment of abdominal organs. Proc AIP Conf. 2010;1215:239–242.
9. Quesson B, et al. Real-time volumetric MRI thermometry of focused ultrasound ablation in vivo: A feasibility study in pig liver and kidney. NMR Biomed. 2011;24:145–153. [PubMed]
10. Ries M, de Senneville BD, Regard Y, Moonen C. Combined magnetic resonance imaging and ultrasound echography guidance for motion compensated HIFU interventions. Proc AIP Conf. 2012;1503:202–206.
11. Meaney PM, Zhou T, Fanning MW, Geimer SD, Paulsen KD. Microwave thermal imaging of scanned focused ultrasound heating: Phantom results. Int J Hyperthermia. 2008;24:523–536. [PMC free article] [PubMed]
12. Rivens I, Shaw A, Civale J, Morris H. Treatment monitoring and thermometry for therapeutic focused ultrasound. Int J Hyperthermia. 2007 Mar;23:121–139. [PubMed]
13. Straube WL, Arthur RM. Theoretical estimation of the temperature dependence of backscattered ultrasonic power for noninvasive thermometry. Ultrasound Med Biol. 1994;20:915–922. [PubMed]
14. Ebbini ES, Yao H, Shrestha A. Dual-mode ultrasound phased arrays for image-guided surgery. Ultrason Imag. 2006 Apr;28:65–82. [PubMed]
15. Lewis MA, Staruch RM, Chopra R. Thermometry and ablation monitoring with ultrasound. Int J Hyperthermia. 2015;31(2):163–181. [PMC free article] [PubMed]
16. Seip R, VanBaren P, Cain CA, Ebbini ES. Noninvasive real-time multipoint temperature control for ultrasound phased array treatments. IEEE Trans Ultrason Ferroelectr Freq Control. 1996 Nov;43(6):1063–1073.
17. Civale J, et al. Calibration of ultrasound backscatter temperature imaging for high-intensity focused ultrasound treatment planning. Ultrasound Med Biol. 2013;39:1596–1612. [PubMed]
18. Seip R, Ebbini ES. Noninvasive estimation of tissue temperature response to heating fields using diagnostic ultrasound. IEEE Trans Biomed Eng. 1995 Aug;42(8):828–839. [PubMed]
19. Amini AN, Ebbini ES, Georgiou TT. Noninvasive estimation of tissue temperature via high-resolution spectral analysis techniques. IEEE Trans Biomed Eng. 2005 Feb;52(2):221–228. [PubMed]
20. Arthur RM, Trobaugh JW, Straube WL, Moros EG. Temperature dependence of ultrasonic backscattered energy in motion-compensated images. IEEE Trans Ultrason Ferroelectr Freq Control. 2005 Oct;52(10):1644–1652. [PubMed]
21. Shishitani T, Matsuzawa R, Yoshizawa S, Umemura S-I. Changes in backscatter of liver tissue due to thermal coagulation induced by focused ultrasound. J Acoust Soc Amer. 2013;134:1724–1730. [PubMed]
22. Zhong H, Wan MX, Jiang YF, Wang SP. Monitoring imaging of lesions induced by high intensity focused ultrasound based on differential ultrasonic attenuation and integrated backscatter estimation. Ultrasound Med Biol. 2007;33:82–94. [PubMed]
23. Casper A, Liu D, Ebbini ES. Realtime control of multiple-focus phased array heating patterns based on noninvasive ultrasound thermography. IEEE Trans Biomed Eng. 2012 Jan;59(1):95–105. [PMC free article] [PubMed]
24. Bayat M, Ballard JR, Ebbini ES. In vivo ultrasound thermography in presence of temperature heterogeneity and natural motions. IEEE Trans Biomed Eng. 2015 Feb;62(2):450–457. [PMC free article] [PubMed]
25. Foiret J, Ferrara KW. Spatial and temporal control of hyperthermia using real time ultrasonic thermal strain imaging with motion compensation. PLoS One. 2015;10(8):e0134938. [PMC free article] [PubMed]
26. Wirtzfeld LA, et al. Techniques and evaluation from a cross-platform imaging comparison of quantitative ultrasound parameters in an in vivo rodent fibroadenoma model. IEEE Trans Ultrason Ferroelectr Freq Control. 2013 Jul;60(7):1386–1400. [PMC free article] [PubMed]
27. Ghoshal G, Lavarello RJ, Kemmerer JP, Miller RJ, Oelze ML. Ex vivo study of quantitative ultrasound parameters in fatty rabbit livers. Ultrasound Med Biol. 2012 Oct 10;38:2238–2248. [PMC free article] [PubMed]
28. Oelze ML, Zachary JF, O’Brien WD., Jr Parametric imaging of rat mammary tumors in vivo for the purposes of tissue characterization. J Ultrasound Med. 2002 Nov;21:1201–1210. [PubMed]
29. Ghoshal G, et al. Quantitative ultrasound imaging for monitoring in situ high-intensity focused ultrasound exposure. Ultrason Imag. 2014 Oct;36:239–255. [PMC free article] [PubMed]
30. Ghoshal G, Luchies AC, Blue JP, Oelze ML. Temperature dependent ultrasonic characterization of biological media. J Acoust Soc Amer. 2011 Oct;130:2203–2211. [PubMed]
31. Kemmerer JP, Oelze ML. Ultrasonic assessment of thermal therapy in rat liver. Ultrasound Med Biol. 2012 Dec;38:2130–2137. [PMC free article] [PubMed]
32. Ghoshal G, Oelze ML. Use of quantitative ultrasound to detect temperature variations in biological phantoms due to heating,” in. Proc IEEE Ultrason Symp. 2009:1780–1783.
33. Morris H, Rivens I, Shaw A, Ter Haar G. Investigation of the viscous heating artefact arising from the use of thermocouples in a focused ultrasound field. Phys Med Biol. 2008;53(17):4759–4776. [PubMed]
34. Oelze ML, O’Brien WD., Jr Defining optimal axial and lateral resolution for estimating scatterer properties from volumes using ultrasound backscatter. J Acoust Soc Amer. 2004;115:3226–3234. [PubMed]
35. Yao LX, Zagzebski JA, Madsen EL. Backscatter coefficient measurements using a reference phantom to extract depth-dependent instrumentation factors. Ultrason Imag. 1990;12:58–70. [PubMed]
36. Oelze ML, Zachary JF, O’Brien WD., Jr Characterization of tissue microstructure using ultrasonic backscatter: Theory and technique for optimization using a Gaussian form factor. J Acoust Soc Amer. 2002 Sep;112:1202–1211. [PubMed]
37. Madsen EL, et al. Interlaboratory comparison of ultrasonic backscatter, attenuation, and speed measurements. J Ultrasound Med. 1999 Sep;18:615–631. [PubMed]
38. Lizzi FL, Astor M, Liu T, Deng C, Coleman DJ, Silverman RH. Ultrasonic spectrum analysis for tissue assays and therapy evaluation. Int J Imag Syst Technol. 1997;8:3–10.
39. Ghoshal G, Kemmerer JP, Oelze ML. Changes in quantitative ultrasound parameters during HIFU application. Proc AIP Conf. 2012;1481:106–111.
40. Kemmerer JP, Ghoshal G, Rubert N, Samimi K, Oelze ML. Quantitative imaging of temperature elevations in tisssues due to thermal therapies. Proc. IEEE Int. Ultrason. Symp; Chicago, IL, USA. 2014. pp. 2153–2156.
41. Bush NL, Rivens I, Ter Haar GR, Bamber JC. Acoustic properties of lesions generated with an ultrasound therapy system. Ultrasound Med Biol. 1993;19:789–801. [PubMed]
42. Gertner MR, Wilson BC, Sherar MD. Ultrasound properties of liver tissue during heating. Ultrasound Med Biol. 1997;23:1395–1403. [PubMed]
43. Kemmererer JP, Ghoshal G, Karanakaran C, Oelze ML. Assessment of high-intensity focused ultrasound treatment of rodent mammary tumors using ultrasound backscatter coefficients. J Acoust Soc Amer. 2013;134(2):1559–1568. [PubMed]