{"id":"W4414489233","doi":"10.21203/rs.3.rs-7578570/v1","title":"Computer Assisted Verbal Autopsy: Comparing Large Language Models to Physicians for Assigning Causes to 6939 Deaths in Sierra Leone from 2019-2022","year":2025,"lang":"en","type":"preprint","venue":"Research Square","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"St. Michael's Hospital","funders":"Bill and Melinda Gates Foundation","keywords":"Sierra leone; Concordance; Verbal autopsy; Epidemiology; Population; Public health; Health care","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":[0.005571267,0.0007656735,0.0004180968,0.001213343,0.0004087409,0.0007763326,0.0008186221,0.0007689954,0.00154878],"category_scores_gemma":[0.01932251,0.0003245717,0.0008693152,0.0005959697,0.0003993339,0.0007244052,0.0007689316,0.0009026276,0.000577718],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008346973,"about_ca_system_score_gemma":0.001532348,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.04684569,"about_ca_topic_score_gemma":0.05192857,"domain_scores_codex":[0.9982994,0.001189154,0.00008536455,0.0002780211,0.00007589172,0.00007219008],"domain_scores_gemma":[0.9864118,0.01196002,0.0004034014,0.0005224145,0.0004879186,0.0002144497],"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":[0.006212582,0.001354091,0.7303863,0.0004230952,0.003254544,0.0005787244,0.001206403,0.1292314,0.001323883,0.0006809431,0.01108365,0.1142642],"study_design_scores_gemma":[0.001062912,0.001673786,0.3700804,0.0003536956,0.001888184,0.0005960984,0.003670859,0.6105612,0.002173941,0.003791848,0.004016997,0.0001300861],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.992779,0.000566232,0.002335539,0.0007636522,0.00005851371,0.00005021848,0.002735239,0.0001742256,0.0005374802],"genre_scores_gemma":[0.9895768,0.0002485445,0.002625435,0.0001093503,0.00003751377,0.00004487233,0.006657213,0.00003154105,0.0006686554],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.04684569,"threshold_uncertainty_score":0.09314609,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1005446132672882,"score_gpt":0.431635171643048,"score_spread":0.3310905583757597,"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."}}