{"id":"W4390745503","doi":"10.1038/s41746-023-00970-0","title":"Large language models to identify social determinants of health in electronic health records","year":2024,"lang":"en","type":"article","venue":"npj Digital Medicine","topic":"Chronic Disease Management Strategies","field":"Medicine","cited_by":254,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"U.S. National Library of Medicine; European Commission; ViewRay; NRG Oncology; Radiological Society of North America; National Cancer Institute; National Institutes of Health; Conquer Cancer Foundation; U.S. Department of Health and Human Services","keywords":"Social determinants of health; Health care; Medicine; Political science; Law","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006570422,0.002210509,0.001069644,0.001682408,0.0009186429,0.001426301,0.002303034,0.001837875,0.003843245],"category_scores_gemma":[0.01874647,0.0009168813,0.002770868,0.001225809,0.0006209163,0.002254214,0.00168702,0.004341784,0.002226254],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001587533,"about_ca_system_score_gemma":0.001644612,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02500094,"about_ca_topic_score_gemma":0.03830668,"domain_scores_codex":[0.9973153,0.001692656,0.0001524549,0.0005670348,0.0001254403,0.0001470795],"domain_scores_gemma":[0.9841273,0.01441792,0.0003339811,0.0003919326,0.0005863669,0.0001424808],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001378183,0.0009810225,0.03624057,0.000706387,0.0008586143,0.0007398472,0.00125933,0.6635621,0.00260119,0.005954942,0.03063195,0.2550858],"study_design_scores_gemma":[0.00005713351,0.0000669166,0.001461333,0.00004677104,0.00006209726,0.00004724426,0.00009804532,0.9908628,0.0004180076,0.005618887,0.001236896,0.00002389022],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4735448,0.006142838,0.4736991,0.007634806,0.0008915791,0.0007323984,0.01657634,0.01587977,0.004898289],"genre_scores_gemma":[0.814564,0.0007251628,0.1514015,0.001916307,0.0003774178,0.0009507653,0.02370046,0.0005315059,0.005832843],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02500094,"threshold_uncertainty_score":0.04971081,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04766063648794992,"score_gpt":0.4148725473884519,"score_spread":0.3672119109005019,"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."}}