Reproduce “BrainIB: Interpretable Brain Network-based Psychiatric Diagnosis with Graph Information Bottleneck” 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
Kaizhong Zheng, Shujian Yu, Baojuan Li, Robert Jenssen, Badong Chen
BrainIB classifies diagnostic groups from resting-state fMRI by selecting a compact set of functional connections that retains label-predictive information. It selects edges and uses matrix-based Rényi entropy to train a subgraph representation. On REST-meta-MDD, ten-fold accuracy was 0.700 ± 0.022, versus 0.668 ± 0.040 for CI-GNN; held-out-site results were not best at every site and are not prospective clinical validation.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.