{"id":"W4388591773","doi":"10.1001/jamacardio.2023.4859","title":"Natural Language Processing for Adjudication of Heart Failure in a Multicenter Clinical Trial","year":2023,"lang":"en","type":"letter","venue":"JAMA Cardiology","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":36,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Heart, Lung, and Blood Institute","keywords":"Adjudication; Medicine; Heart failure; Concordance; Clinical trial; Medical record; Gold standard (test); Cohort; Intensive care medicine; Emergency medicine; Pediatrics; Internal medicine","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["research_integrity"],"consensus_categories":[],"category_scores_codex":[0.001728717,0.0002176261,0.0009906021,0.0003042177,0.00004440569,0.00004600337,0.0009035991,0.001271795,0.000001368874],"category_scores_gemma":[0.002439275,0.0001991989,0.0003756198,0.0003072382,0.00009323981,0.0001255408,0.0002654432,0.002751361,0.00001872846],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007526379,"about_ca_system_score_gemma":0.0002432433,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001315798,"about_ca_topic_score_gemma":0.00004781672,"domain_scores_codex":[0.9963349,0.001254167,0.0009106325,0.0007783077,0.0002535961,0.0004684216],"domain_scores_gemma":[0.9969285,0.001772222,0.000474471,0.000591905,0.0001949451,0.00003796958],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001010549,0.00001829973,0.006812894,0.0009569567,0.00006523859,0.0001229685,0.0006492872,0.00008197279,0.00001693581,0.00004140431,0.9329214,0.05730216],"study_design_scores_gemma":[0.02545978,0.000867089,0.01976496,0.0004586134,0.00004761532,0.00006656373,0.00007524575,0.07125775,0.000006148556,0.0002181588,0.8812187,0.0005593331],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"commentary","genre_gemma":"empirical","genre_scores_codex":[0.01041634,0.0009038571,0.01262825,0.9686777,0.004699429,0.002290245,0.0000470598,0.0002913074,0.00004579178],"genre_scores_gemma":[0.4946817,0.00003127045,0.06805786,0.4029542,0.03156709,0.001095955,0.0008797782,0.0001584587,0.0005736634],"genre_candidate":"commentary","genre_consensus":null,"teacher_disagreement_score":0.5657235,"threshold_uncertainty_score":0.9995493,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04726143691355651,"score_gpt":0.3963823846668722,"score_spread":0.3491209477533157,"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."}}