OpenBrain: An Auditable Confidence-Aware Release of 35,838 Whole-Brain MRI Parcellations

Forum topic · 小凯 · 2026-11-04

Summary

OpenBrain is a confidence-aware release of 35,838 provenance-verified OpenNeuro T1w MRI cases drawn from 607 source datasets, introduced in arXiv paper 2611.05348 (Qizhen Lan et al., 2026). Rather than a directory of standalone segmentation files, OpenBrain organizes generated whole-brain MRI parcellations as auditable case records: each record links a primary parcellation, two committee parcellations, source and license provenance, raw QC-7 measurements, percentile risks, a combined risk score R, and a confidence tier. QC-7 is a reference-free reliability panel covering label geometry, image-label consistency, and committee disagreement, letting users filter, weight, audit, or re-threshold generated labels without rerunning inference. Under a frozen evaluation protocol, lower QC-7 risk correlates with higher generated-label quality across external cohorts, and fixed-budget training with low-risk selection improves paired macro Dice by +3.2 to +3.7 percentage points over random selection. The release includes raw axes, committee outputs, provenance, and datasheet documentation. Dataset: https://huggingface.co/datasets/lanqz7766/OpenBrain; Code: https://github.com/lanqz7766/openbrain.

Overview

Field: Computer Vision
Authors: Qizhen Lan, Yu-Chun Hsu, Yuxiang Wei, Lijing Zhu, Zenan Sun, Liang He et al.
Published: 2026-11-03
arXiv: 2611.05348

Abstract

Automatic tools can now generate whole-brain MRI parcellations at cohort scale, but generated-label releases often provide little evidence about which cases are reliable enough to reuse. We introduce OpenBrain, a confidence-aware release of 35,838 provenance-verified OpenNeuro T1w cases from 607 source datasets.

OpenBrain is organized as an auditable case-record release rather than a directory of standalone segmentation files: each record links a primary parcellation, two committee parcellations, source/license provenance, raw QC-7 measurements, percentile risks, a combined risk score R, and a confidence tier.

QC-7 is a reference-free reliability panel over label geometry, image-label consistency, and committee disagreement, so users can filter, weight, audit, or re-threshold generated labels without rerunning inference.

Evaluation

The released evidence is evaluated under a frozen protocol:

Across external cohorts, lower QC-7 risk corresponds to higher generated-label quality.
In fixed-budget training, lower-risk selections improve paired macro Dice over random selection by +3.2 to +3.7 percentage points.

OpenBrain also exposes raw axes, committee outputs, provenance, and datasheet documentation, so confidence claims remain inspectable after download.

Scope

OpenBrain does not turn generated labels into manual annotations or clinical-grade segmentations. Its contribution is a large open generated-parcellation substrate, a reusable case-level risk-evidence layer, and an external validation protocol for release-time QC.