Electronic　medical　records　(EMRs)　support　the　development　of　machine　learning　algorithms　for　predicting　disease　incidence,　patient　response　to　treatment,　and　other　healthcare　events.　But　so　far　most　algorithms　have　been　centralized,　taking　little　account　of　the　decentralized,　non-identically　independently　distributed　(non-IID),　and　privacy-sensitive　characteristics　of　EMRs　that　can　complicate　data　collection,　sharing　and　learning.　To　address　this　challenge,　we　introduced　a　community-based　federated　machine　learning　(CBFL)　algorithm　and　evaluated　it　on　non-IID　ICU　EMRs.　Our　algorithm　clustered　the　distributed　data　into　clinically　meaningful　communities　that　captured　similar　diagnoses　and　geographical　locations,　and　learnt　one　model　for　each　community.　Throughout　the　learning　process,　the　data　was　kept　local　at　hospitals,　while　locally-computed　results　were　aggregated　on　a　server.　Evaluation　results　show　that　CBFL　outperformed　the　baseline　federated　machine　learning　(FL)　algorithm　in　terms　of　Area　Under　the　Receiver　Operating　Characteristic　Curve　(ROC　AUC),　Area　Under　the　Precision-Recall　Curve　(PR　AUC),　and　communication　cost　between　hospitals　and　the　server.　Furthermore,　communities＇　performance　difference　could　be　explained　by　how　dissimilar　one　community　was　to　others.