Determine from CT data whether a tumor will be controlled by definitive radiation therapy

This is a dataset for a competition, which was organized by our radiation oncology team, led by Assistant Professor Clifton D. Fuller, University of Texas MD Anderson Cancer Center (MDACC), as a MICCAI grand challenge. Contestants were tasked to predict, using expert-segmented contrast-enhanced computed tomography (CT) images, primary local recurrence via participant-developed radiomics workflows based on MDACC provided dataset of anonymized DICOM files that represents a relatively uniform cohort of 288 oropharynx cancer patients, supplemented with relevant clinical data, known etiological/biological correlates (specifically, human papilloma virus "HPV" status) as ground truth. We'd already provided the contestants with a defined “Training” cohort as a "prior" dataset that includes all input and outcome data, to build up an algorithm.