Reproduce “Learning the Hierarchical Organization in Brain Network for Brain Disorder Diagnosis” 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

Jingfeng Tang, Peng Cao, Guangqi Wen, Jinzhu Yang, Xiaoli Liu, Osmar R. Zaiane

A framework called Brain Hierarchical Organization Learning (BrainHO) was developed to dynamically identify hierarchical organization in brain networks for diagnosing neurological disorders. It achieved state-of-the-art classification accuracy on Autism Spectrum Disorder (ASD) and Major Depressive Disorder (MDD) datasets by learning intrinsic functional sub-networks, including those spanning traditional boundaries.

Key insights

  • Learning the intrinsic hierarchical organization of brain networks, driven by functional features, provides a more accurate and flexible representation than relying on fixed functional atlases.
  • Dynamically formed sub-networks can capture critical cross-network interactions that are often missed by atlas-dependent methods, revealing more comprehensive disease-related biomarkers.
  • Orthogonality constraints and hierarchical consistency strategies are vital for ensuring that learned multi-scale representations are diverse, non-redundant, and semantically aligned, enhancing both performance and interpretability.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.