{"id":"W4381093987","doi":"10.2196/46267","title":"Comparing Natural Language Processing and Structured Medical Data to Develop a Computable Phenotype for Patients Hospitalized Due to COVID-19: Retrospective Analysis","year":2023,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Medicine; Coronavirus disease 2019 (COVID-19); Medical record; Retrospective cohort study; Health records; Electronic health record; Emergency medicine; Medical emergency; Pediatrics; Health care; Internal medicine; Disease","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0121556,0.0005179027,0.0005987635,0.004645879,0.0002926733,0.00140596,0.0006285084,0.000511237,0.0005552843],"category_scores_gemma":[0.04860224,0.0003378982,0.001293444,0.003399993,0.0007806836,0.001014027,0.001054256,0.0007218961,0.0001946208],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009197575,"about_ca_system_score_gemma":0.001372942,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004602607,"about_ca_topic_score_gemma":0.003418184,"domain_scores_codex":[0.9902019,0.004679142,0.001865993,0.001473501,0.001346717,0.0004327133],"domain_scores_gemma":[0.9326699,0.03970619,0.01490918,0.004414366,0.007427946,0.0008725561],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0003970794,0.0001111652,0.993955,0.0000710571,0.0001934742,0.000104129,0.0002339973,0.0009911752,0.0002018894,0.00009567141,0.0001985519,0.003446773],"study_design_scores_gemma":[0.0000783195,0.001167185,0.9639255,0.0001313615,0.0003435604,0.0006576877,0.001628372,0.02888769,0.001708313,0.0004325575,0.0009858609,0.00005350408],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9937277,0.0001721268,0.003269659,0.00006541258,0.000007021263,0.0001917624,0.002379185,0.00002033364,0.0001667509],"genre_scores_gemma":[0.9925452,0.00007061823,0.003447556,0.00004184135,0.00001013599,0.0002065692,0.003626246,0.00001205634,0.00003970795],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0121556,"threshold_uncertainty_score":0.06428576,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03037505858305085,"score_gpt":0.373488774862108,"score_spread":0.3431137162790571,"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."}}