Given that many users engage in polysubstance use, Mount Sinai et al introduced an explainable, connectivity-based multiclass classification framework that models distributed brain network organization to delineate both shared and substance-specific mechanisms of addiction including cannabis and alcohol use disorders.


