Lack of online range verification limits efficacy of particle therapy for many tumor sites. Real-time thermoacoustic range verification could enable more aggressive treatment planning and hypofractionation for tumor sites that can be imaged with ultrasound during treatment. Our objective was to experimentally demonstrate accuracy and robustness of thermoacoustic range estimates relative to ultrasound images despite acoustic heterogeneity and discrepancies between assumed and true soundspeed. Prior results were for weak acoustic scatterers with known soundspeeds. 250 ns pulses of 0.26 Gy of 16 MeV protons and 2.3 Gy of 60 MeV helium ions were delivered to water and oil targets, respectively. Thermoacoustic signals with DC-4 MHz bandwidth were detected by a 96-channel ultrasound array placed 6-10 cm distal to the Bragg peak. One-way beamforming with an assumed soundspeed was performed to estimate range. The same soundspeed and transducer array were used to generate ultrasound images via two-way ultrasound beamforming. An air gap phantom displaced the Bragg peak by 6.5 mm to demonstrate accuracy. The scanner’s soundspeed setting was altered by ±5% to demonstrate robustness to soundspeed errors. Tissue mimicking gelatin and a 5 mm thick bone sample were introduced to demonstrate robustness to acoustic heterogeneity. Single ion pulse measurements sufficed during the helium run, but signal averaging was required for protons. Estimates of the entry point agreed with the air-target interface in ultrasound images and range estimates agreed with Monte Carlo simulations to within 300 μm microns, even when TA emissions traveled through a strong acoustic scatterer. Estimated Bragg peak locations were translated 6.5 mm by the air gap phantom and correctly identified scenarios when the beam stopped inside bone, but did not accurately estimate range in bone. Soundspeed errors dilate and acoustic heterogeneities deform ultrasound images. Thermoacoustic range estimates are transformed similarly and are robust relative to ultrasound images of underlying anatomy. When the target can be visualized with ultrasound during treatment, thermoacoustic range verification may enable real-time motion management and range verification. Therapeutic systems deliver higher (200+ MeV) energy protons using pulse durations exceeding 5 μs and generate thermoacoustic emissions with sonar bandwidths (DC-100 kHz). Therefore, custom acoustic hardware will be required to detect low frequency thermoacoustic emissions and also generate high-resolution ultrasound images.Abstract 3713; Table 1Experimentwaterwater & bonewater & air gapsafflower oilIonprotonprotonproton4HeBeam Energy (MeV)16.0±0.1516.0±0.1516.0±0.1560.7±0.4Bragg curve FWHM (μm)390240390230distal HWTM (μm)1609016080Monte Carlo Range (mm)2.471.948.952.51Thermoacoustic Range (mm)2.65±0.091.36±0.616.59±0.042.76±0.04N8887 Open table in a new tab

Purpose: To demonstrate that very high frequency (VHF) induced thermoacoustics has the potential to provide quantitative images of electrical conductivity in Siemens/meter, much as shear wave elastography provides tissue stiffness in kPa. Quantitatively imaging a large organ requires exciting thermoacoustic pulses throughout the volume and broadband detection of those pulses because tomographic image reconstruction preserves frequency content. Applying the half-wavelength limit to a 200-micron inclusion inside a 7.5 cm diameter organ requires measurement sensitivity to frequencies ranging from 4 MHz down to 10 kHz, respectively. VHF irradiation provides superior depth penetration over near infrared used in photoacoustics. Additionally, VHF signal production is proportional to electrical conductivity, and prostate cancer is known to suppress electrical conductivity of prostatic fluid. Methods: A dual-transducer system utilizing a P4-1 array connected to a Verasonics V1 system augmented by a lower frequency focused single element transducer was developed. Simultaneous acquisition of VHF-induced thermoacoustic pulses by both transducers enabled comparison of transducer performance. Data from the clinical array generated a stack of 96-images with separation of 0.3 mm, whereas the single element transducer imaged only in a single plane. In-plane resolution and quantitative accuracy were measured at isocenter. Results: The array provided volumetric imaging capability with superior resolution whereas the single element transducer provided superior quantitative accuracy. Combining axial images from both transducers preserved resolution of the P4-1 array and improved image contrast. Neither transducer was sensitive to frequencies below 50 kHz, resulting in a DC offset and low-frequency shading over fields of view exceeding 15 mm. Fresh human prostates were imaged ex vivo and volumetric reconstructions reveal structures rarely seen in diagnostic images. Conclusion: Quantitative whole-organ thermoacoustic tomography will be feasible by sparsely interspersing transducer elements sensitive to the low end of the ultrasonic range.

Purpose: The potential of particle therapy has not yet been fully realized due to inaccuracies in range verification. The purpose of this work was to correlate the Bragg peak location with target structure, by overlaying thermoacoustic localization of the Bragg peak onto an ultrasound image. Methods: Pulsed delivery of 50 MeV protons was accomplished by a fast chopper installed between the ion source and the inflector of the 88″ cyclotron at Lawrence Berkeley National Lab. 2 Gy were delivered in 2 µs by a beam with peak current of 2 µA. Thermoacoustic emissions were detected by a cardiac array and Verasonics V1 ultrasound system, which also generated a grayscale ultrasound image. 1024 thermoacoustic pulses were averaged before filtering and one-way beamforming focused signal onto the Bragg peak location with perfect co-registration to the ultrasound images. Data was collected in a room temperature water bath and gelatin phantom with a cavity designed to mimic the intestine, in which gas pockets can displace the Bragg peak. Experiments were performed with the cavity both empty and filled with olive oil. Results: In the waterbath overlays of the Bragg peak agreed with Monte Carlo simulations to within 800±170 µm. Agreement within 1.3 ± 0.2 mm was achieved in the gelatin phantom, although relative stopping powers were estimated only to first order from CT scans. Protoacoustic signals were detected after travel from the Bragg peak through 29 mm and 65 mm of phantom material when the cavity was empty and full of olive oil, respectively. Conclusion: Protoacoustic range verification is feasible with a commercial clinical ultrasound array, but at doses exceeding the clinical realm. Further optimization of both transducer array and injection line chopper is required to enable range verification within a 2 Gy dose limit, which would enable online adaptive treatment. This work was supported in part by a UWM Intramural Instrumentation Grant and by the Director, Office of Science, Office of Nuclear Physics, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. YMQ was supported by a UWM-OUR summer fellowship.