Overview

Schema Version

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

Name

UCSF Adult Longitudinal Post-Treatment Diffuse Glioma (UCSF-ALPTDG) MRI Dataset

Link

https://imagingdatasets.ucsf.edu/dataset/2

Indexing

Keywords: Diffuse Glioma, Neuro-Oncology, Resection Cavity

Content: MR, NR, OI

RadLex: RID6434, RID4026

Author(s)

Brandon K. K. Fields

Evan Calabrese

John Mongan

Soonmee Cha

Christopher P. Hess

Leo P. Sugrue

Susan M. Chang

Tracy L. Luks

Javier E. Villanueva-Meyer

Andreas M. Rauschecker

Jeffrey D. Rudie

Organization(s)

University of California San Francisco

Duke University Medical Center

License

Text: Non-commercial use with data use agreement

URL: https://imagingdatasets.ucsf.edu/dataset/2

Contact

Jeff.Rudie@gmail.com

Ethical review

Retrospective, single-center, Health Insurance Portability and Accountability Act–compliant study with UCSF Institutional Review Board approval and waiver for informed consent.

Comments

This expertly annotated, anonymized dataset consists of multimodal three-dimensional (3D) MR images of diffuse gliomas at two consecutive posttreatment time points and is readily available for training and validation of novel and existing artificial intelligence workflows. The dataset includes corresponding clinical history and expert voxelwise annotations.

Dataset

Motivation

To support large-scale machine learning validations and bolster efforts to develop artificial intelligence techniques for longitudinal disease monitoring in high-risk patient populations.

Sampling

Retrospective, single-center study. 298 patients with diffuse glioma and two consecutive imaging time points were identified from institutional imaging archives (mPower; Nuance Communications). Patients underwent posttreatment surveillance MRI between January 2018 and December 2019.

Partitioning scheme

The dataset was partitioned into a random training set of 248 patients and a test set of 50 patients.

Missing information

Derived from a single institution, protocol, scanner model, and field strength, which may limit generalizability. Only a single reference standard voxelwise annotation is available, limiting interrater reliability assessments. Heterogeneity in follow-up intervals between scans may limit comparisons between patients.

Relationships between instances

Each patient includes multimodal 3D MR images at two consecutive posttreatment time points, allowing for longitudinal change analysis.

Confidentiality

Anonymized dataset, Health Insurance Portability and Accountability Act–compliant.

Re-identification

Anonymized, implying no re-identification is possible.