{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.001440368,0.0002204454,0.0005848561,0.0004605971,0.0003216217,0.0002624105,0.00198568,0.0001755725,0.00002993493],"category_scores_gemma":[0.00982192,0.0001843961,0.00003758639,0.00398172,0.00006467388,0.0005040106,0.002562588,0.0004937795,0.00002117142],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001892628,"about_ca_system_score_gemma":0.0008527206,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001273041,"about_ca_topic_score_gemma":0.0001764895,"domain_scores_codex":[0.9964293,0.00008397004,0.0007638424,0.0004112343,0.001743274,0.0005683731],"domain_scores_gemma":[0.9969326,0.000346505,0.0002149522,0.0006995643,0.0003551975,0.001451231],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000183977,0.0001906805,0.3341135,0.003743002,0.0006245736,0.0001369879,0.272695,0.003323602,0.000001590487,0.002523359,0.04045197,0.3420117],"study_design_scores_gemma":[0.0007472944,0.00009244283,0.1243893,0.00008457316,0.0000219867,0.000006910104,0.0003788574,0.8714904,6.277002e-7,0.0000694892,0.00249978,0.0002183793],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4884894,0.00003768411,0.5028977,0.006363108,0.0003678548,0.001177174,0.0000562338,0.0005535983,0.00005722732],"genre_scores_gemma":[0.8837733,0.000002188861,0.1075867,0.007889095,0.00009175052,0.00007333959,0.0005497875,0.00001478524,0.00001902637],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8681667,"threshold_uncertainty_score":0.9985188,"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."}}