DNN Surrogate Towards Fast Ablation Zone Prediction in Thermal Ablation
摘要
Numerical models of thermal ablation can be extremely successful in predicting the ablation zone after a percutaneous cancer treatment. However, such models are often too slow for real-time forward, let alone inverse, treatment planning. This paper explores the feasibility of a deep neural network (DNN) surrogate approach to address this issue without sacrificing accuracy.