ConnSearch: A framework for functional connectivity analysis designed for interpretability and effectiveness at limited sample sizes

Paul C. Bogdan, Alexandru D. Iordan, Jonathan Shobrook, Florin Dolcos

Abstract: Highlights•Novel machine learning approach effective for even limited datasets (N = 25–50).•Accurate classifiers fit using just small fractions (0.4%) of the connectome.•Includes subject-specific techniques that account for across-brain heterogeneity.•ConnSearch yields highly interpretable outputs consistent with existing literature.