This study moves glioma recurrence assessment from lesion-level labels toward spatial maps of suspected recurrent tumor. Using physiologic MRI and machine learning, the authors linked predicted intralesional patterns…