Reproduce “OpenBrain: An Auditable Generated-Label Release for Whole-Brain MRI Parcellation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Qizhen Lan, Yu-Chun Hsu, Yuxiang Wei, Lijing Zhu, Zenan Sun, Liang He, Lishan Yu, Xiaoqian Jiang
OpenBrain releases auditable whole-brain MRI parcellation records for 35,838 provenance-verified cases from 607 datasets. Each record pairs a primary generated label with two committee outputs, provenance, seven quality checks, risk ranks, and a confidence tier. At 25% retention, low-risk selection improved macro Dice, including 1.5 percentage points on manual-label Mindboggle-101; scores rank evidence, not error probabilities or clinical guarantees.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.