Abstract: Disclosed herein are methods, non-transitory computer readable media, and systems for subphenotyping acute respiratory distress syndrome (ARDS) patients by analyzing electronic health data (EHR) using a subphenotype classifier. According to their classification, different treatments can be selected which are likely to be efficacious in treating ARDS. Such methods, non-transitory computer readable media, and systems are useful for rapid classification and guided treatment in critical care settings, such as in hospitals.

Abstract: A method for determining at least one of a medical diagnosis recommendation and a medical intervention recommendation for a subject. At least one of electronic health record (EHR) data and biomarker data for the subject are input into a diagnostic/intervention recommendation model that comprises parameters and a function. The parameters are identified based on a training dataset that comprises a plurality of training samples. Each training sample is associated with a retrospective subject and includes at least one of EHR data and biomarker data for the retrospective subject. The function represents a relation between the at least one of EHR data and biomarker data for the subject received as inputs to the diagnostic/intervention recommendation model, and at least one of a medical diagnosis recommendation and intervention recommendation for the subject generated as an output of the model.