Abstract: Techniques for detecting real-time proton beam dose include retrieving conversion data for converting proton dose to acoustic pressure based on medical imagery. 3D medical imagery is retrieved of a first region of interest (ROI) inside a first subject. A real-time signal indicates ultrasound pressure at a 2D ultrasound array disposed outside and in contact with the first subject close to the first ROI. Real-time 3D dose delivered in the first ROI is determined by inputting the real-time signal into a dose detection module that includes a neural network. The neural network is trained on multiple instances of a training set. Each instance includes: simulated proton dose delivered to a similar ROI inside a second subject; simulated pressure generation from that simulated dose based on the conversion data and a similar 3D medical image; and simulated acoustic transmission from the similar ROI to a similar two-dimensional ultrasound transducer array similarly disposed.

Abstract: The present embodiments relate generally to increasing the precision of radiotherapy by measuring the absolute dose delivered to the tumor and surrounding normal tissue during the treatment. More particularly, some embodiments relate to an imaging reconstruction system for X-ray-induced acoustic computed tomography (XACT) using a model-based reconstruction method. Instead of reconstructing relative dose information for radiation beam localization, the system is capable of reconstructing absolute in vivo dose information. The XACT absolute in vivo dosimetry tool holds great potential for personalized cancer treatment and better outcomes. In some embodiments, thermal parameters, such as Gruneisen parameters, are used to convert reconstructed pressure information to dose. In addition, to avoid problems caused by electrical system gain, calibration tools, such as ion chambers, can be used to calibrate the system.

Abstract: The present embodiments relate generally to increasing the precision of radiotherapy by measuring the absolute dose delivered to the tumor and surrounding normal tissue during the treatment. More particularly, some embodiments relate to an imaging reconstruction system for X-ray-induced acoustic computed tomography (XACT) using a model-based reconstruction method. Instead of reconstructing relative dose information for radiation beam localization, the system is capable of reconstructing absolute in vivo dose information. The XACT absolute in vivo dosimetry tool holds great potential for personalized cancer treatment and better outcomes. In some embodiments, thermal parameters, such as Gruneisen parameters, are used to convert reconstructed pressure information to dose. In addition, to avoid problems caused by electrical system gain, calibration tools, such as ion chambers, can be used to calibrate the system.

Abstract: Techniques for detecting real-time proton beam dose include retrieving conversion data for converting proton dose to acoustic pressure based on medical imagery. 3D medical imagery is retrieved of a first region of interest (ROI) inside a first subject. A real-time signal indicates ultrasound pressure at a 2D ultrasound array disposed outside and in contact with the first subject close to the first ROI. Real-time 3D dose delivered in the first ROI is determined by inputting the real-time signal into a dose detection module that includes a neural network. The neural network is trained on multiple instances of a training set. Each instance includes: simulated proton dose delivered to a similar ROI inside a second subject; simulated pressure generation from that simulated dose based on the conversion data and a similar 3D medical image; and simulated acoustic transmission from the similar ROI to a similar two-dimensional ultrasound transducer array similarly disposed.