FITBIR

FITBIR is a data repository for traumatic brain injury research, built by the NIH Center for Information Technology on its BRICS platform together with NINDS and the US Department of Defense. Studies upload clinical assessments, imaging, biomarker and omics data using the NINDS common data elements, and a global unique identifier (GUID) links records of the same participant across studies. Researchers query and download data across all shared studies through one data access request. This entry describes the repository's imaging holdings as one dataset, because access, agreement and data format are the same for every study in it.

Composition

In January 2024 the repository held 184 studies with data for more than 101,000 subjects, including more than 295,000 clinical image datasets. From 24 August 2026 FITBIR hosts data from NIH-funded studies only, and data from other funders is no longer available there. The public statistics for 8 October 2026 list 43 studies, 25 of them shared. In the public study list, 24 of the 43 studies report imaging data, 16 of which are shared. Studies range from large observational cohorts and clinical trials (TRACK-TBI, ProTECT III, ADAPT) to small concussion and sports studies, in adults and children. Some studies include controls. A subject total for the current holdings is not published, and per-study counts cannot simply be added because the GUID can join the same person across studies.

Acquisition

Each study used its own scanners and protocol; there is no common imaging protocol. Images are submitted with metadata in imaging form structures. On 8 October 2026 these were used by the following numbers of studies: MR 17, diffusion 15, functional MR 9, CT 6, MEG 2, PET 1 and digital pathology 1. Submitted images are checked for duplicates, unnecessary images, identifier inconsistencies and possible personal information.

Annotations

There are no repository-wide labels. Six studies submitted structured image reads in a common imaging read form.

Known limitations

• Access needs a data access request signed by a senior investigator and the institutional signing official, approved by a Data Access Committee for one year and for one non-commercial project.
• Content changes as studies are added or withdrawn, and shrank sharply in August 2026 when non-NIH data left the repository. Counts depend on the date.
• Studies differ in design, population, imaging timing and protocol, so pooled analyses need harmonisation.
• Some studies are private and become shared only after their sharing date.