Radiation therapy planning using deep convolutional neural network

摘要

PROBLEM TO BE SOLVED: To provide a radiation therapy plan for a patient by training a deep convolutional neural network. Training involves collecting patient data based on at least one image of the patient's anatomical structure from the patient, and determining a treatment plan, including a set of control points, from the collected patient data. It involves training a deep convolutional neural network for regression using the determined treatment plan and the corresponding collected patient data to determine a treatment plan containing control points from the collected patient data. .. The trained model can be used, for example, to provide a radiation therapy plan in real time. [Selection diagram] FIG. 9A