{"id":"W3118996476","doi":"10.2196/24207","title":"Federated Learning of Electronic Health Records to Improve Mortality Prediction in Hospitalized Patients With COVID-19: Machine Learning Approach","year":2021,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":211,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Center for Advancing Translational Sciences","keywords":"Machine learning; Artificial intelligence; Lasso (programming language); Computer science; Logistic regression; Generalizability theory; Health records; Multilayer perceptron; Predictive modelling; Deep learning; Electronic health record; Health care; Artificial neural network; Statistics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001892889,0.0002420985,0.0005375631,0.0002232458,0.0002511624,0.00009855992,0.0004791382,0.0001978199,0.00006693904],"category_scores_gemma":[0.002097124,0.0002095435,0.00005732744,0.001360932,0.00005886731,0.000438725,0.0003559044,0.001779896,0.000007816302],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000569745,"about_ca_system_score_gemma":0.002808928,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008109197,"about_ca_topic_score_gemma":0.0002171662,"domain_scores_codex":[0.99562,0.0006306219,0.001256842,0.0003417462,0.00144362,0.0007072138],"domain_scores_gemma":[0.9977359,0.000167436,0.0006091835,0.0003708344,0.0002853054,0.0008313299],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00008009739,0.0004417449,0.92623,0.001825508,0.00005749016,0.00001000813,0.01675447,0.01346942,0.000001243598,0.001389234,0.0002347273,0.03950612],"study_design_scores_gemma":[0.002095596,0.001506024,0.09360258,0.0001714916,0.000004617181,0.00001287573,0.00053052,0.8984975,0.000008311243,0.00005656392,0.003280792,0.0002331143],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6007665,0.00006186117,0.3960352,0.001390027,0.0001433923,0.0008028501,0.000009229083,0.0002570757,0.0005337837],"genre_scores_gemma":[0.9826149,0.00003804366,0.01438789,0.002347361,0.00003412112,0.0001188615,0.0003286792,0.00002154445,0.0001085783],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8850281,"threshold_uncertainty_score":0.8544935,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01285349720121259,"score_gpt":0.3090294637525138,"score_spread":0.2961759665513012,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}