William M. Whelan

Optical and acoustic properties at 1064 nm of polyvinyl chloride-plastisol for use as a tissue phantom in biomedical optoacoustics

A novel optoacoustic phantom made of polyvinyl chloride-plastisol (PVCP) for optoacoustic studies is described. The optical and acoustic properties of PVCP were measured. Titanium dioxide (TiO2) powder and black plastic colour (BPC) were used to introduce scattering and absorption, respectively, in the phantoms. The optical absorption coefficient (mua) at 1064 nm was determined using an optoacoustic method, while diffuse reflectance measurements were used to obtain the optical reduced scattering coefficient (mus). These optical properties were calculated to be mua = (12.818 +/- 0.001)ABPC cm(-1) and mus = (2.6 +/- 0.2)S(TiO2) + (1.4 +/- 0.1) cm(-1), where ABPC is the BPC per cent volume concentration, and S(TiO2) is the TiO2 volume concentration (mg mL(-1)). The speed of sound in PVCP was measured to be (1.40 +/- 0.02) x 10(3) m s(-1) using the pulse echo transmit receive method, with an acoustic attenuation of (0.56 +/- 1.01) f(1.51+/-0.06)MHz (dB cm(-1)) in the frequency range of 0.61-1.25 MHz, and a density, calculated by measuring the displacement of water, of 1.00 +/- 0.04 g cm(-3). The speed of sound and density of PVCP are similar to tissue, and together with the user-adjustable optical properties, make this material well suited for developing tissue-equivalent phantoms for biomedical optoacoustics.

Study of laser-induced thermoelastic deformation of native and coagulated ex-vivo bovine liver tissues for estimating their optical and thermomechanical properties

Several studies have explored the potential of optoacoustic imaging for monitoring thermal therapies, yet the origin of the contrast in the images is not well understood. A technique is required to measure the changes in the optical and thermomechanical properties of tissues upon coagulation to better understand this contrast. An interferometric method is presented for measuring simultaneously the optical and thermomechanical properties of native and coagulated ex-vivo bovine tissue samples based on analysis of the surface displacement of irradiated samples. Surface displacements are measured after irradiation by short laser pulses at 750 nm. A 51% decrease in the optical attenuation depth is observed for coagulated liver samples compared to native samples. No significant differences in the Grüneisen coefficient are measured in the native and coagulated tissue samples. A mean value of 0.12 for the Grüneisen coefficient is measured for both native and coagulated liver tissues. The displacement profiles exhibit consistent differences between the two tissue types. To assess the changes in the sample mechanical properties, the experimental data also are compared to numerical solutions of the equation for thermoelastic deformation. The results demonstrate that differences in the tissue expansion dynamics arise from higher values of elastic modulus for coagulated liver samples compared to native ones.

Dynamic modeling of interstitial laser photocoagulation: Implications for lesion formation in liver in vivo

Interstitial Laser Photocoagulation (ILP) is a minimally invasive cancer treatment technique, whereby optical energy from implanted optical fibers is used to therapeutically heat small, solid tumors. In this work, the potential of ILP without tissue charring is investigated.

Determination of the optical properties of turbid media using relative interstitial radiance measurements: Monte Carlo study, experimental validation, and sensitivity analysis.

Interstitial quantification of the optical properties of tissue is important in biomedicine for both treatment planning of minimally invasive laser therapies and optical spectroscopic characterization of tissues, for example, prostate cancer. In a previous study, we analyzed a method first demonstrated by Dickey et al., [Phys. Med. Biol. 46, 2359 (2001)] to utilize relative interstitial steady-state radiance measurements for recovering the optical properties of turbid media. The uniqueness of point radiance measurements were demonstrated in a forward sense, and strategies were suggested for improving performance under noisy experimental conditions. In this work, we test our previous conclusions by fitting the P3 approximation for radiance to Monte Carlo predictions and experimental data in tissue-simulating phantoms. Fits are performed at: 1. a single sensor position (0.5 or 1 cm), 2. two sensor positions (0.5 and 1 cm), and 3. a single sensor position (0.5 or 1 cm) with input knowledge of the samples effective attenuation coefficient. The results demonstrate that single sensor radiance measurements can be used to retrieve optical properties to within approximately 20%, provided the transport albedo is greater than approximately 0.9. Furthermore, compared to the single sensor fits, employing radiance data at two sensor positions did not significantly improve the accuracy of recovered optical properties. However, with knowledge of the effective attenuation coefficient of the medium, optical properties can be retrieved experimentally to within approximately 10% for an albedo greater or equal to 0.5.

Optoacoustic characterization of prostate cancer in an in vivo transgenic murine model

Abstract. Optoacoustic (OA) imaging was employed to distinguish normal from neoplastic tissues in a transgenic murine model of prostate cancer. OA images of five tumor-bearing mice and five age-matched controls across a 14  mm×14  mm region of interest (ROI) on the lower abdomen were acquired using a reverse-mode OA imaging system (Seno Medical Instruments Inc., San Antonio, Texas). Neoplastic prostate tissue was identified based on the OA signal amplitude in combination with spectral analysis of the OA radio frequency (RF) data. Integration of the signal amplitude images was performed to construct two-dimensional images of the ROI. The prostate tumors generated higher amplitude signals than those of the surrounding tissues, with contrast ratios ranging from 31 to 36 dB. The RF spectrum analysis showed significant differences between the tumor and the control mice. The midband fit was higher by 5 dB (62%), the intercept higher by 4 dB (57%) and the spectral slope higher by 0.4  dB/MHz (50%) for neoplastic prostate tissue compared to normal tissues in the control mice. The results demonstrate that OA offers high contrast imaging of prostate cancer in vivo.

Models and measurements of light intensity changes during laser interstitial thermal therapy: implications for optical monitoring of the coagulation boundary location

We have developed a multi-region spherical Monte Carlo (MC) model to simulate the dynamic changes in light intensity measured during laser interstitial thermal therapy (LITT). Model predictions were validated experimentally in tissue-simulating albumen phantoms with well-characterized optical properties that vary dynamically with LITT in a way similar to tissue. For long treatments (2.5 W, approximately 1800 s), the transient light intensity changes demonstrated better qualitative agreement with a three-region MC model (with an inner layer of fully coagulated optical properties, a middle layer of partially coagulated properties and an outer region of native properties); for short treatments (4 W, approximately 240 s), better qualitative agreement was seen with a two-region MC model (with an inner layer of fully coagulated properties and outer region of native properties). These differences were attributed to differences in coagulation formation during low- and high-powered heating regimes, respectively. At the end of heating, a three-region coagulation zone was observed for both heating schemes. Quantitatively, final light intensity changes at the end of heating were compared with changes predicted by both two- and three-region MC for the same experimentally measured coagulation size and found to agree within approximately 30% for both models. The developed MC model helps lend insight into the nature of thermal coagulation events occurring for low and high power LITT irradiation schemes.

Changes in relative light fluence measured during laser heating: implications for optical monitoring and modelling of interstitial laser photocoagulation.

Dynamic changes in internal light fluence were measured during interstitial laser heating of tissue phantoms and ex vivo bovine liver. In albumen phantoms, the results demonstrate an unexpected rise in optical power transmitted approximately I cm away from the source during laser exposure at low power (0.5-1 W), and a decrease at higher powers (1.5-2.5 W) due to coagulation and possibly charring. Similar trends were observed in liver tissue, with a rise in interstitial fluence observed during 0.5 W exposure and a drop in interstitial fluence seen at higher powers (1-1.5 W) due to tissue coagulation. At 1.5 W irradiation an additional, later decrease was also seen which was most likely due to tissue charring. Independent spectrophotometric studies in Naphthol Green dye indicate the rise in fluence observed in the heated albumen phantoms may have been primarily due to light exposure causing photobleaching of the absorbing chromophore. and not due to heat effects. Experiments in liver tissue demonstrated that the observed rise in fluence is dependent on the starting temperature of the tissue. Correlating changes in light fluence with key clinical endpoints/events such as the onset of tissue coagulation or charring may be useful for on-line monitoring and control of laser thermal therapy via interstitial fluence sensors.

Separation of absorption and scattering properties of turbid media using relative spectrally resolved cw radiance measurements

We present a new method for extracting the effective attenuation coefficient and the diffusion coefficient from relative spectrally resolved cw radiance measurements using the diffusion approximation. The method is validated on both simulated and experimental radiance data sets using Intralipid-1% as a test platform. The effective attenuation coefficient is determined from a simple algebraic expression constructed from a ratio of two radiance measurements at two different source–detector separations and the same 90° angle. The diffusion coefficient is determined from another ratio constructed from two radiance measurements at two angles (0° and 180°) and the same source–detector separation. The conditions of the validity of the method as well as possible practical applications are discussed.

Radiance-based monitoring of the extent of tissue coagulation during laser interstitial thermal therapy

Optical monitoring relates the dynamic changes in measured light intensity to the extent of treatment-induced coagulation that occurs during laser interstitial thermal therapy. We utilized a two-region Monte Carlo simulation to elucidate the nature of the changes in interstitial radiance and fluence that result from the formation of a volume of thermal coagulation surrounding a cylindrical emitter. Using simulation results, we demonstrate that radiance sensors are more sensitive than traditional fluence sensors to coagulation-induced scattering changes. Radiance measurements take advantage of directional detection angles that are more receptive to the onset and passing of the coagulation boundary. We performed experiments with albumen phantoms to demonstrate the practicality of the radiance method for monitoring interstitial laser thermal therapy.

Development and testing of an optoacoustic imaging system for monitoring and guiding prostate cancer therapies

Laser Optoacoustic Imaging System (LOIS) combines high tissue contrast based on the optical properties of tissue and high spatial resolution based on ultrawide-band ultrasonic detection. Patients undergoing thermal or photodynamic therapy of prostate cancer may benefit from capability of LOIS to detect and monitor treatment-induced changes in tissue optical properties and blood flow. The performance of a prototype LOIS was evaluated via 2D optoacoustic images of dye-colored objects of various shapes, small tubes with blood simulating veins and arteries, and thermally coagulated portions of chicken breasts imbedded tissue-mimicking gelatin phantoms. The optoacoustic image contrast was proportional to the ratio of the absorption coefficient between the embedded objects and the surrounding gel. The contrast of the venous blood relative to the background exceeded 250%, and the contrast of the thermally coagulated portions of flesh relative to the untreated tissue ranged between -100% to +200%, dependent on the optical wavelength. We used a 32-element optoacoustic transducer array and a novel design of low-noise preamplifiers and wide-band amplifiers to perform these studies. The system was optimized for imaging at a depth of ~50 mm. The system spatial resolution was better than 1-mm. The advantages and limitations of various signal-processing methods were investigated. LOIS demonstrates clinical potential for non- or minimally-invasive monitoring of treatment-induced tissue changes.