Fig. 3From: Radiomics based targeted radiotherapy planning (Rad-TRaP): a computational framework for prostate cancer treatment planning with MRIQuantitative results of the voxel-wise predictions using the radiomics trained machine learning classifier in terms of AUCs for individual patients from D 1. The classifier was trained on T2w, ADC MRI sequences and T2w alone to show that misalignment between T2w and ADC MRI affects the performance of the classifier (patients 2, 5 and 6)Back to article page