Abstract: A computer system receives, from a user, a set of clinical eligibility criteria. The computer system retrieves, from a database, medical records for a plurality of subjects. The computer system inputs, into a model comprising 1000 parameters: information corresponding to the set of clinical eligibility criteria; and information corresponding to the medical records for at least a first subset of the plurality of subjects. The computer system receives, from the model, responsive to the inputting, a ranking of the first subset of the plurality of subjects indicative of relative relevance of the respective medical histories of the plurality of subjects to the set of clinical eligibility criteria. The computer system selects, based on the ranking, a second subset of the plurality of subjects for inclusion in the cohort.