RESUMO
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.
Assuntos
Aprendizado Profundo , Terapia com Prótons , Masculino , Humanos , Prótons , Imageamento Tridimensional , Próstata , Processamento de Imagem Assistida por Computador/métodosRESUMO
BACKGROUND: Protoacoustic (PA) imaging has the potential to provide real-time 3D dose verification of proton therapy. However, PA images are susceptible to severe distortion due to limited angle acquisition. Our previous studies showed the potential of using deep learning to enhance PA images. As the model was trained using a limited number of patients' data, its efficacy was limited when applied to individual patients. PURPOSE: In this study, we developed a patient-specific deep learning method for protoacoustic imaging to improve the reconstruction quality of protoacoustic imaging and the accuracy of dose verification for individual patients. METHODS: Our method consists of two stages: in the first stage, a group model is trained from a diverse training set containing all patients, where a novel deep learning network is employed to directly reconstruct the initial pressure maps from the radiofrequency (RF) signals; in the second stage, we apply transfer learning on the pre-trained group model using patient-specific dataset derived from a novel data augmentation method to tune it into a patient-specific model. Raw PA signals were simulated based on computed tomography (CT) images and the pressure map derived from the planned dose. The reconstructed PA images were evaluated against the ground truth by using the root mean squared errors (RMSE), structural similarity index measure (SSIM) and gamma index on 10 specific prostate cancer patients. The significance level was evaluated by t-test with the p-value threshold of 0.05 compared with the results from the group model. RESULTS: The patient-specific model achieved an average RMSE of 0.014 ( p < 0.05 ${{{p}}}<{0.05}$ ), and an average SSIM of 0.981 ( p < 0.05 ${{{p}}}<{0.05}$ ), out-performing the group model. Qualitative results also demonstrated that our patient-specific approach acquired better imaging quality with more details reconstructed when comparing with the group model. Dose verification achieved an average RMSE of 0.011 ( p < 0.05 ${{{p}}}<{0.05}$ ), and an average SSIM of 0.995 ( p < 0.05 ${{{p}}}<{0.05}$ ). Gamma index evaluation demonstrated a high agreement (97.4% [ p < 0.05 ${{{p}}}<{0.05}$ ] and 97.9% [ p < 0.05 ${{{p}}}<{0.05}$ ] for 1%/3 and 1%/5 mm) between the predicted and the ground truth dose maps. Our approach approximately took 6 s to reconstruct PA images for each patient, demonstrating its feasibility for online 3D dose verification for prostate proton therapy. CONCLUSIONS: Our method demonstrated the feasibility of achieving 3D high-precision PA-based dose verification using patient-specific deep-learning approaches, which can potentially be used to guide the treatment to mitigate the impact of range uncertainty and improve the precision. Further studies are needed to validate the clinical impact of the technique.
RESUMO
BACKGROUND: Fast low angle shot hyperfractionation (FLASH) radiotherapy (RT) holds promise for improving treatment outcomes and reducing side effects but poses challenges in radiation delivery accuracy due to its ultra-high dose rates. This necessitates the development of novel imaging and verification technologies tailored to these conditions. PURPOSE: Our study explores the effectiveness of proton-induced acoustic imaging (PAI) in tracking the Bragg peak in three dimensions and in real time during FLASH proton irradiations, offering a method for volumetric beam imaging at both conventional and FLASH dose rates. METHODS: We developed a three-dimensional (3D) PAI technique using a 256-element ultrasound detector array for FLASH dose rate proton beams. In the study, we tested protoacoustic signal with a beamline of a FLASH-capable synchrocyclotron, setting the distal 90% of the Bragg peak around 35 mm away from the ultrasound array. This configuration allowed us to assess various total proton radiation doses, maintaining a consistent beam output of 21 pC/pulse. We also explored a spectrum of dose rates, from 15 Gy/s up to a FLASH rate of 48 Gy/s, by administering a set number of pulses. Furthermore, we implemented a three-dot scanning beam approach to observe the distinct movements of individual Bragg peaks using PAI. All these procedures utilized a proton beam energy of 180 MeV to achieve the maximum possible dose rate. RESULTS: Our findings indicate a strong linear relationship between protoacoustic signal amplitudes and delivered doses (R2 = 0.9997), with a consistent fit across different dose rates. The technique successfully provided 3D renderings of Bragg peaks at FLASH rates, validated through absolute Gamma index values. CONCLUSIONS: The protoacoustic system demonstrates effectiveness in 3D visualization and tracking of the Bragg peak during FLASH proton therapy, representing a notable advancement in proton therapy quality assurance. This method promises enhancements in protoacoustic image guidance and real-time dosimetry, paving the way for more accurate and effective treatments in ultra-high dose rate therapy environments.
RESUMO
Objective. The primary goal of this research is to demonstrate the feasibility of radiation-induced acoustic imaging (RAI) as a volumetric dosimetry tool for ultra-high dose rate FLASH electron radiotherapy (FLASH-RT) in real time. This technology aims to improve patient outcomes by accurate measurements ofin vivodose delivery to target tumor volumes.Approach. The study utilized the FLASH-capable eRT6 LINAC to deliver electron beams under various doses (1.2 Gy pulse-1to 4.95 Gy pulse-1) and instantaneous dose rates (1.55 × 105Gy s-1to 2.75 × 106Gy s-1), for imaging the beam in water and in a rabbit cadaver with RAI. A custom 256-element matrix ultrasound array was employed for real-time, volumetric (4D) imaging of individual pulses. This allowed for the exploration of dose linearity by varying the dose per pulse and analyzing the results through signal processing and image reconstruction in RAI.Main Results. By varying the dose per pulse through changes in source-to-surface distance, a direct correlation was established between the peak-to-peak amplitudes of pressure waves captured by the RAI system and the radiochromic film dose measurements. This correlation demonstrated dose rate linearity, including in the FLASH regime, without any saturation even at an instantaneous dose rate up to 2.75 × 106Gy s-1. Further, the use of the 2D matrix array enabled 4D tracking of FLASH electron beam dose distributions on animal tissue for the first time.Significance. This research successfully shows that 4Din vivodosimetry is feasible during FLASH-RT using a RAI system. It allows for precise spatial (â¼mm) and temporal (25 frames s-1) monitoring of individual FLASH beamlets during delivery. This advancement is crucial for the clinical translation of FLASH-RT as enhancing the accuracy of dose delivery to the target volume the safety and efficacy of radiotherapeutic procedures will be improved.
Assuntos
Elétrons , Animais , Coelhos , Dosagem Radioterapêutica , Radiometria/métodos , Acústica , Dosimetria in Vivo/métodosRESUMO
X-ray-induced acoustic (XA) computerized tomography (XACT) is an evolving imaging technique that aims to reconstruct the X-ray energy deposition from XA measurements. Main challenges in XACT are the poor signal-to-noise ratio and limited field-of-view, which cause artifacts in the images. We demonstrate the efficacy of model-based (MB) algorithms for three-dimensional XACT and compare with the traditional algorithms. The MB algorithm is based on iterative, matrix-free approach for regularized-least-squares minimization corresponding to XACT. The matrix-free-LSQR (MF-LSQR) and the non-iterative model-backprojection (MBP) reconstructions were evaluated and compared with universal backprojection (UBP), time-reversal (TR) and fast-Fourier transform (FFT)-based reconstructions for numerical and experimental XACT datasets. The results demonstrate the capability of MF-LSQR algorithm to reduce noisy artifacts thus yielding better reconstructions. MBP and MF-LSQR algorithms perform particularly well with the experimental XACT dataset, where noise in signals significantly affects the reconstruction of the target in UBP and FFT-based reconstructions. The TR reconstruction for experimental XACT are comparable to MF-LSQR, but takes thrice as much time and filters the frequency components greater than maximum frequency supported by the grid, resulting loss of resolution. The MB algorithms are able to overcome the challenges in XACT and hence are vital for the clinical translation of XACT.
RESUMO
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 significantly impairs its accuracy for dose verification. In this study, we developed a deep learning method with a Recon- Enhance two-stage strategy for protoacoustic imaging to address the limited view issue. Specifically, in the Recon-stage, a transformer-based network was developed to reconstruct initial pressure maps from radiofrequency 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. The results evaluated on a dataset of 126 prostate cancer patients achieved an average RMSE of 0.0292, and an average SSIM of 0.9618, significantly 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 seconds, demonstrating its feasibility for online 3D dose verification for prostate proton therapy.
RESUMO
Purpose: High-precision radiation therapy is crucial for cancer treatment. Currently, the delivered dose can only be verified via simulations with phantoms, and an in-tumor, online dose verification is still unavailable. An innovative detection method called x-ray-induced acoustic computed tomography (XACT) has recently shown the potential for imaging the delivered radiation dose within the tumor. Prior XACT imaging systems have required tens to hundreds of signal averages to achieve high-quality dose images within the patient, which reduces its real-time capability. Here, we demonstrate that XACT dose images can be reproduced from a single x-ray pulse (4 µs) with sub-mGy sensitivity from a clinical linear accelerator. Methods and Materials: By immersing an acoustic transducer in a homogeneous medium, it is possible to detect pressure waves generated by the pulsed radiation from a clinical linear accelerator. After rotating the collimator, signals of different angles are obtained to perform a tomographic reconstruction of the dose field. Using 2-stage amplification with further bandpass filtering increases the signal-to-noise ratio (SNR). Results: Acoustic peak SNR and voltage values were recorded for singular and dual-amplifying stages. The SNR for single-pulse mode was able to satisfy the Rose criterion, and the collected signals were able to reconstruct 2-dimensional images from the 2 homogeneous media. Conclusions: By overcoming the low SNR and requirement of signal averaging, single-pulse XACT imaging holds great potential for personalized dose monitoring from each individual pulse during radiation therapy.
RESUMO
Radiation-induced acoustic (RA) imaging is a promising technique for visualizing the invisible radiation energy deposition in tissues, enabling new imaging modalities and real-time therapy monitoring. However, RA imaging signal often suffers from poor signal-to-noise ratios (SNRs), thus requiring measuring hundreds or even thousands of frames for averaging to achieve satisfactory quality. This repetitive measurement increases ionizing radiation dose and degrades the temporal resolution of RA imaging, limiting its clinical utility. In this study, we developed a general deep inception convolutional neural network (GDI-CNN) to denoise RA signals to substantially reduce the number of frames needed for averaging. The network employs convolutions with multiple dilations in each inception block, allowing it to encode and decode signal features with varying temporal characteristics. This design generalizes GDI-CNN to denoise acoustic signals resulting from different radiation sources. The performance of the proposed method was evaluated using experimental data of x-ray-induced acoustic, protoacoustic, and electroacoustic signals both qualitatively and quantitatively. Results demonstrated the effectiveness of GDI-CNN: it achieved x-ray-induced acoustic image quality comparable to 750-frame-averaged results using only 10-frame-averaged measurements, reducing the imaging dose of x-ray-acoustic computed tomography (XACT) by 98.7%; it realized proton range accuracy parallel to 1500-frame-averaged results using only 20-frame-averaged measurements, improving the range verification frequency in proton therapy from 0.5 to 37.5 Hz; it reached electroacoustic image quality comparable to 750-frame-averaged results using only a single frame signal, increasing the electric field monitoring frequency from 1 fps to 1k fps. Compared to lowpass filter-based denoising, the proposed method demonstrated considerably lower mean-squared-errors, higher peak-SNR, and higher structural similarities with respect to the corresponding high-frame-averaged measurements. The proposed deep learning-based denoising framework is a generalized method for few-frame-averaged acoustic signal denoising, which significantly improves the RA imaging's clinical utilities for low-dose imaging and real-time therapy monitoring.
Assuntos
Aprendizado Profundo , Redes Neurais de Computação , Tomografia Computadorizada por Raios X/métodos , Razão Sinal-Ruído , Acústica , Processamento de Imagem Assistida por Computador/métodosRESUMO
Radiation-induced acoustic (RA) imaging is a promising technique for visualizing radiation energy deposition in tissues, enabling new imaging modalities and real-time therapy monitoring. However, it requires measuring hundreds or even thousands of averages to achieve satisfactory signal-to-noise ratios (SNRs). This repetitive measurement increases ionizing radiation dose and degrades the temporal resolution of RA imaging, limiting its clinical utility. In this study, we developed a general deep inception convolutional neural network (GDI-CNN) to denoise RA signals to substantially reduce the number of averages. The multi-dilation convolutions in the network allow for encoding and decoding signal features with varying temporal characteristics, making the network generalizable to signals from different radiation sources. The proposed method was evaluated using experimental data of X-ray-induced acoustic, protoacoustic, and electroacoustic signals, qualitatively and quantitatively. Results demonstrated the effectiveness and generalizability of GDI-CNN: for all the enrolled RA modalities, GDI-CNN achieved comparable SNRs to the fully-averaged signals using less than 2% of the averages, significantly reducing imaging dose and improving temporal resolution. The proposed deep learning framework is a general method for few-frame-averaged acoustic signal denoising, which significantly improves RA imaging's clinical utilities for low-dose imaging and real-time therapy monitoring.
RESUMO
BACKGROUND: Radiation dosimetry is essential for radiation therapy (RT) to ensure that radiation dose is accurately delivered to the tumor. Despite its wide use in clinical intervention, the delivered radiation dose can only be planned and verified via simulation. This makes precision radiotherapy challenging while in-line verification of the delivered dose is still absent in the clinic. X-ray-induced acoustic computed tomography (XACT) has recently been proposed as an imaging tool for in vivo dosimetry. PURPOSE: Most of the XACT studies focus on localizing the radiation beam. However, it has not been studied for its potential for quantitative dosimetry. The aim of this study was to investigate the feasibility of using XACT for quantitative in vivo dose reconstruction during radiotherapy. METHODS: Varian Eclipse system was used to generate simulated uniform and wedged 3D radiation field with a size of 4 cm × $ \times \ $ 4 cm. In order to use XACT for quantitative dosimetry measurements, we have deconvoluted the effects of both the x-ray pulse shape and the finite frequency response of the ultrasound detector. We developed a model-based image reconstruction algorithm to quantify radiation dose in vivo using XACT imaging, and universal back-projection (UBP) reconstruction is used as comparison. The reconstructed dose was calibrated before comparing it to the percent depth dose (PDD) profile. Structural similarity index matrix (SSIM) and root mean squared error (RMSE) are used for numeric evaluation. Experimental signals were acquired from 4 cm × $ \times \ $ 4 cm radiation field created by Linear Accelerator (LINAC) at depths of 6, 8, and 10 cm beneath the water surface. The acquired signals were processed before reconstruction to achieve accurate results. RESULTS: Applying model-based reconstruction algorithm with non-negative constraints successfully reconstructed accurate radiation dose in 3D simulation study. The reconstructed dose matches well with the PDD profile after calibration in experiments. The SSIMs between the model-based reconstructions and initial doses are over 85%, and the RMSEs of model-based reconstructions are eight times lower than the UBP reconstructions. We have also shown that XACT images can be displayed as pseudo-color maps of acoustic intensity, which correspond to different radiation doses in the clinic. CONCLUSION: Our results show that the XACT imaging by model-based reconstruction algorithm is considerably more accurate than the dose reconstructed by UBP algorithm. With proper calibration, XACT is potentially applicable to the clinic for quantitative in vivo dosimetry across a wide range of radiation modalities. In addition, XACT's capability of real-time, volumetric dose imaging seems well-suited for the emerging field of ultrahigh dose rate "FLASH" radiotherapy.
Assuntos
Dosimetria in Vivo , Raios X , Tomografia Computadorizada por Raios X , Radiometria/métodos , Imagens de Fantasmas , Acústica , Dosagem RadioterapêuticaRESUMO
Dose delivery uncertainty is a major concern in proton therapy, adversely affecting the treatment precision and outcome. Recently, a promising technique, proton-acoustic (PA) imaging, has been developed to provide real-timein vivo3D dose verification. However, its dosimetry accuracy is limited due to the limited-angle view of the ultrasound transducer. In this study, we developed a deep learning-based method to address the limited-view issue in the PA reconstruction. A deep cascaded convolutional neural network (DC-CNN) was proposed to reconstruct 3D high-quality radiation-induced pressures using PA signals detected by a matrix array, and then derive precise 3D dosimetry from pressures for dose verification in proton therapy. To validate its performance, we collected 81 prostate cancer patients' proton therapy treatment plans. Dose was calculated using the commercial software RayStation and was normalized to the maximum dose. The PA simulation was performed using the open-source k-wave package. A matrix ultrasound array with 64 × 64 sensors and 500 kHz central frequency was simulated near the perineum to acquire radiofrequency (RF) signals during dose delivery. For realistic acoustic simulations, tissue heterogeneity and attenuation were considered, and Gaussian white noise was added to the acquired RF signals. The proposed DC-CNN was trained on 204 samples from 69 patients and tested on 26 samples from 12 other patients. Predicted 3D pressures and dose maps were compared against the ground truth qualitatively and quantitatively using root-mean-squared-error (RMSE), gamma-index (GI), and dice coefficient of isodose lines. Results demonstrated that the proposed method considerably improved the limited-view PA image quality, reconstructing pressures with clear and accurate structures and deriving doses with a high agreement with the ground truth. Quantitatively, the pressure accuracy achieved an RMSE of 0.061, and the dose accuracy achieved an RMSE of 0.044, GI (3%/3 mm) of 93.71%, and 90%-isodose line dice of 0.922. The proposed method demonstrates the feasibility of achieving high-quality quantitative 3D dosimetry in PA imaging using a matrix array, which potentially enables the online 3D dose verification for prostate proton therapy.
Assuntos
Aprendizado Profundo , Terapia com Prótons , Masculino , Humanos , Terapia com Prótons/métodos , Prótons , Próstata , Acústica , Imagens de FantasmasRESUMO
BACKGROUND: We previously developed a deep learning model to augment the quality of four-dimensional (4D) cone-beam computed tomography (CBCT). However, the model was trained using group data, and thus was not optimized for individual patients. Consequently, the augmented images could not depict small anatomical structures, such as lung vessels. METHODS: In the present study, the transfer learning method was used to further improve the performance of the deep learning model for individual patients. Specifically, a U-Net-based model was first trained to augment 4D-CBCT using group data. Next, transfer learning was used to fine tune the model based on a specific patient's available data to improve its performance for that individual patient. Two types of transfer learning were studied: layer-freezing and whole-network fine-tuning. The performance of the transfer learning model was evaluated by comparing the augmented CBCT images with the ground truth images both qualitatively and quantitatively using a structure similarity index matrix (SSIM) and peak signal-to-noise ratio (PSNR). The results were also compared to those obtained using only the U-Net method. RESULTS: Qualitatively, the patient-specific model recovered more detailed information of the lung area than the group-based U-Net model. Quantitatively, the SSIM improved from 0.924 to 0.958, and the PSNR improved from 33.77 to 38.42 for the whole volumetric images for the group-based U-Net and patient-specific models, respectively. The layer-freezing method was found to be more efficient than the whole-network fine-tuning method, and had a training time as short as 10 minutes. The effect of augmentation by transfer learning increased as the number of projections used for CBCT reconstruction decreased. CONCLUSIONS: Overall, the patient-specific model optimized by transfer learning was efficient and effective at improving image qualities of augmented undersampled three-dimensional (3D)- and 4D-CBCT images, and could be extremely valuable for applications in image-guided radiation therapy.