45 A comparison of pseudo-CT generation methods for prostate MRI-based dose planning: deep learning, patch-based, atlas-based and bulk-density methods

Abstract

Introduction. Deep learning methods (DLM) have recently been developed to generate pseudo-CT (pCT) from MRI for radiotherapy dose calculation. The main advantage of these methods is the speed of pCT generation. This study aims to evaluate the accuracy of two DLMs proposed to generate pCT from MRI in prostate radiotherapy: a generative adversarial network (GAN) DLM using a perceptual loss, the classical U-Net DLM and compare them with three current state-of-the art Methods. a patch-based method (PBM), an atlas-based method (ABM) and a bulk density method (BDM).