https://atlas.rsna.org/schemas/2025-11/dataset.json

UCSF Preoperative Diffuse Glioma MRI Dataset

https://doi.org/10.7937/tcia.bdgf-8v37

evan.calabrese@ucsf.edu

Supported in part by the National Institutes of Health under awards number NCI:U01CA242871 and T32EB001631, as well as by the Radiological Society of North America Research & Education Foundation under award number RR2011.

Data collection was performed in accordance with relevant guidelines and regulations and was approved by the UCSF institutional review board with a waiver for consent.

The UCSF-PDGM dataset includes 501 patients with histopathologically proven diffuse gliomas imaged with a standardized 3-T preoperative brain tumor MRI protocol. It features predominantly three-dimensional imaging, including diffusion and perfusion imaging, IDH mutation status, MGMT promotor methylation status for WHO grade 3 and 4 gliomas, and treatment details including extent of resection and overall survival.

The UCSF-PDGM dataset has been made publicly available in hopes that researchers around the world will use these data to continue to push the boundaries of AI applications for diffuse gliomas.

The dataset population consisted of 501 adult patients with histopathologically confirmed WHO grade 2–4 diffuse gliomas (following the 2021 WHO Classification of Central Nervous System Tumors) who underwent preoperative MRI, initial tumor resection, and tumor genetic testing at a single medical center between 2015 and 2021. Patients with any prior history of brain tumor treatment were excluded; however, prior tumor biopsy was allowed (n = 69 of 501 or 14%).

Race and ethnicity data were not available for the study population. Not all patients were tested for MGMT methylation and 1p/19q codeletion.

Eleven cases were excluded because of severe image artifacts (patient motion or hardware related), and 33 cases were excluded due to one or more missing series.