Feasibility study of 3D time-reversal reconstruction of proton-induced acoustic signals for dose verification in the head and the liver: A simulation study

Technical note: Application of an optical hydrophone to ionoacoustic range detection in a tissue-mimicking agar phantom

BACKGROUND

Ionoacoustics is a promising approach to reduce the range uncertainty in proton therapy. A miniature-sized optical hydrophone (OH) was used as a measuring device to detect weak ionoacoustic signals with a high signal-to-noise ratio in water. However, further development is necessary to prevent wave distortion because of nearby acoustic impedance discontinuities while detection is conducted on the patient's skin.

PURPOSE

A prototype of the probe head attached to an OH was fabricated and the required dimensions were experimentally investigated using a 100-MeV proton beam from a fixed-field alternating gradient accelerator and k-Wave simulations. The beam range of the proton in a tissue-mimicking phantom was estimated by measuring γ-waves and spherical ionoacoustic waves with resonant frequency (SPIRE).

METHODS

Four sizes of probe heads were fabricated from agar blocks for the OH. Using the prototype, the γ-wave was detected at distal and lateral positions to the Bragg peak on the phantom surface for proton beams delivered at seven positions. For SPIRE, independent measurements were performed at distal on- and off-axis positions. The range positions were estimated by solving the linear equation using the sensitive matrix for the γ-wave and linear fitting of the correlation curve for SPIRE; they were compared with those measured using a film.

RESULTS

The first peak of the γ-wave was undistorted with the 3 × 3 × 3-cm3 probe head used at the on-axis and 3-cm off-axis positions. The range positions estimated by the γ-wave agreed with the film-based range in the depth direction (the maximum deviation was 0.7 mm), although a 0.6-2.1 mm deviation was observed in the lateral direction. For SPIRE, the deviation was <1 mm for the two measurement positions.

CONCLUSIONS

The attachment of a relatively small-sized probe head allowed the OH to measure the beam range on the phantom surface.

Hybrid-supervised deep learning for domain transfer 3D protoacoustic image reconstruction

Objective. Protoacoustic imaging showed great promise in providing real-time 3D dose verification of proton therapy. However, the limited acquisition angle in protoacoustic imaging induces severe artifacts, which impairs its accuracy for dose verification. In this study, we developed a hybrid-supervised deep learning method for protoacoustic imaging to address the limited view issue.Approach. We proposed a Recon-Enhance two-stage deep learning method. In the Recon-stage, a transformer-based network was developed to reconstruct initial pressure maps from raw acoustic signals. The network is trained in a hybrid-supervised approach, where it is first trained using supervision by the iteratively reconstructed pressure map and then fine-tuned using transfer learning and self-supervision based on the data fidelity constraint. In the enhance-stage, a 3D U-net is applied to further enhance the image quality with supervision from the ground truth pressure map. The final protoacoustic images are then converted to dose for proton verification.Main results. The results evaluated on a dataset of 126 prostate cancer patients achieved an average root mean squared errors (RMSE) of 0.0292, and an average structural similarity index measure (SSIM) of 0.9618, out-performing related start-of-the-art methods. Qualitative results also demonstrated that our approach addressed the limit-view issue with more details reconstructed. Dose verification achieved an average RMSE of 0.018, and an average SSIM of 0.9891. Gamma index evaluation demonstrated a high agreement (94.7% and 95.7% for 1%/3 mm and 1%/5 mm) between the predicted and the ground truth dose maps. Notably, the processing time was reduced to 6 s, demonstrating its feasibility for online 3D dose verification for prostate proton therapy.Significance. Our study achieved start-of-the-art performance in the challenging task of direct reconstruction from radiofrequency signals, demonstrating the great promise of PA imaging as a highly efficient and accurate tool forinvivo3D proton dose verification to minimize the range uncertainties of proton therapy to improve its precision and outcomes.

Reconstruction of thermoacoustic emission sources induced by proton irradiation using numerical time reversal

Objective.Mapping of dose delivery in proton beam therapy can potentially be performed by analyzing thermoacoustic emissions measured by ultrasound arrays. Here, a method is derived and demonstrated for spatial mapping of thermoacoustic sources using numerical time reversal, simulating re-transmission of measured emissions into the medium.Approach.Spatial distributions of thermoacoustic emission sources are shown to be approximated by the analytic-signal form of the time-reversed acoustic field, evaluated at the time of the initial proton pulse. Given calibration of the array sensitivity and knowledge of tissue properties, this approach approximately reconstructs the acoustic source amplitude, equal to the product of the time derivative of the radiation dose rate, mass density, and Grüneisen parameter. This approach was implemented using two models for acoustic fields of the array elements, one modeling elements as line sources and the other as rectangular radiators. Thermoacoustic source reconstructions employed previously reported measurements of emissions from proton energy deposition in tissue-mimicking phantoms. For a phantom incorporating a bone layer, reconstructions accounted for the higher sound speed in bone. Dependence of reconstruction quality on array aperture size and signal-to-noise ratio was consistent with previous acoustic simulation studies.Main results.Thermoacoustic source distributions were successfully reconstructed from acoustic emissions measured by a linear ultrasound array. Spatial resolution of reconstructions was significantly improved in the azimuthal (array) direction by incorporation of array element diffraction. Source localization agreed well with Monte Carlo simulations of energy deposition, and was improved by incorporating effects of inhomogeneous sound speed.Significance.The presented numerical time reversal approach reconstructs thermoacoustic sources from proton beam radiation, based on straightforward processing of acoustic emissions measured by ultrasound arrays. This approach may be useful for ranging and dosimetry of clinical proton beams, if acoustic emissions of sufficient amplitude and bandwidth can be generated by therapeutic proton sources.

Ionoacoustic application of an optical hydrophone to detect proton beam range in water

BACKGROUND

Proton range uncertainty has been the main factor limiting the ability of proton therapy to concentrate doses to tumors to their full potential. Ionoacoustic (IA) range verification is an approach to reducing this uncertainty by detecting thermoacoustic waves emitted from an irradiated volume immediately following a pulsed proton beam delivery; however, the signal weakness has been an obstacle to its clinical application. To increase the signal-to-noise ratio (SNR) with the conventional piezoelectric hydrophone (PH), the detector-sensitive volume needs to be large, but it could narrow the range of available beam angles and disturb real-time images obtained during beam delivery.

PURPOSE

To prevent this issue, we investigated a millimeter-sized optical hydrophone (OH) that exploits the laser interferometric principle. For two types of IA waves [γ-wave emitted from the Bragg peak (BP) and a spherical IA wave with resonant frequency (SPIRE) emitted from the gold fiducial marker (GM)], comparisons were made with PH in terms of waveforms, SNR, range detection accuracy, and signal intensity robustness against the small detector misalignment, particularly for SPIRE.

METHODS

A 100-MeV proton beam with a 27 ns pulse width and 4 mm beam size was produced using a fixed-field alternating gradient accelerator and was irradiated to the water phantom. The GM was set on the beam's central axis. Acrylic plates of various thicknesses, up to 12 mm, were set in front of the phantoms to shift the proton range. OH was set distal and lateral to the beam, and the range was estimated using the time-of-flight method for γ-wave and by comparing with the calibration data (SPIRE intensity versus the distance between the GM and BP) derived from an IA wave transport simulation for SPIRE. The BP dose per pulse was 0.5-0.6 Gy. To measure the variation in SPIRE amplitude against the hydrophone misalignment, the hydrophone was shifted by ± 2 mm at a maximum in lateral directions.

RESULTS

Despite its small size, OH could detect γ-wave with a higher SNR than the conventional PH (diameter, 29 mm), and a single measurement was sufficient to detect the beam range with a submillimeter accuracy in water. In the SPIRE measurement, OH was far more robust against the detector misalignment than the focused PH (FPH) used in our previous study [5%/mm (OH) versus 80%/mm (FPH)], and the correlation between the measured SPIRE intensity and the distance between the GM and BP agreed well with the simulation results. However, the OH sensitivity was lower than the FPH sensitivity, and about 5.6-Gy dose was required to decrease the intensity variation among measurements to less than 10%.

CONCLUSION

The miniature OH was found to detect weak IA signals produced by proton beams with a BP dose used in hypofractionated regimens. The OH sensitivity improvement at the MHz regime is worth exploring as the next step.

Further investigation of 3D dose verification in proton therapy utilizing acoustic signal, wavelet decomposition and machine learning

Online dose verification in proton therapy is a critical task for quality assurance. We further studied the feasibility of using a wavelet-based machine learning framework to accomplishing that goal in three dimensions, built upon our previous work in 1D. The wavelet decomposition was utilized to extract features of acoustic signals and a bidirectional long-short-term memory (Bi-LSTM) recurrent neural network (RNN) was used. The 3D dose distributions of mono-energetic proton beams (multiple beam energies) inside a 3D CT phantom, were generated using Monte-Carlo simulation. The 3D propagation of acoustic signal was modeled using the k-Wave toolbox. Three different beamlets (i.e. acoustic pathways) were tested, one with its own model. The performance was quantitatively evaluated in terms of mean relative error (MRE) of dose distribution and positioning error of Bragg peak (ΔBP), for two signal-to-noise ratios (SNRs). Due to the lack of experimental data for the time being, two SNR conditions were modeled (SNR = 1 and 5). The model is found to yield good accuracy and noise immunity for all three beamlets. The results exhibit an MRE below 0.6% (without noise) and 1.2% (SNR = 5), andΔBPbelow 1.2 mm (without noise) and 1.3 mm (SNR = 5). For the worst-case scenario (SNR = 1), the MRE andΔBPare below 2.3% and 1.9 mm, respectively. It is encouraging to find out that our model is able to identify the correlation between acoustic waveforms and dose distributions in 3D heterogeneous tissues, as in the 1D case. The work lays a good foundation for us to advance the study and fully validate the feasibility with experimental results.

Investigating the accuracy of co-registered ionoacoustic and ultrasound images in pulsed proton beams

The sharp spatial and temporal dose gradients of pulsed ion beams result in an acoustic emission (ionoacoustics), which can be used to reconstruct the dose distribution from measurements at different positions. The accuracy of range verification from ionoacoustic images measured with an ultrasound linear array configuration is investigated both theoretically and experimentally for monoenergetic proton beams at energies relevant for pre-clinical studies (20 and 22 MeV). The influence of the linear sensor array arrangement (length up to 4 cm and number of elements from 5 to 200) and medium properties on the range estimation accuracy are assessed using time-reversal reconstruction. We show that for an ideal homogeneous case, the ionoacoustic images enable a range verification with a relative error lower than 0.1%, however, with limited lateral dose accuracy. Similar results were obtained experimentally by irradiating a water phantom and taking into account the spatial impulse response (geometry) of the acoustic detector during the reconstruction of pressures obtained by moving laterally a single-element transducer to mimic a linear array configuration. Finally, co-registered ionoacoustic and ultrasound images were investigated using silicone inserts immersed in the water phantom across the proton beam axis. By accounting for the sensor response and speed of sound variations (deduced from co-registration with ultrasound images) the accuracy is improved to a few tens of micrometers (relative error less than to 0.5%), confirming the promise of ongoing developments for ionoacoustic range verification in pre-clinical and clinical proton therapy applications.