{"id":"W4401943674","doi":"10.1109/icdh62654.2024.00031","title":"Enhancing Large Language Models with Human Expertise for Disease Detection in Electronic Health Records","year":2024,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"Canadian Institutes of Health Research","keywords":"Computer science; Health records; Electronic health record; Human disease; Disease; Data science; Natural language processing; Health care; Medicine; Pathology","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"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":[0.007063882,0.001397291,0.0008424525,0.002168686,0.0007744602,0.002004185,0.001655711,0.001349966,0.001807655],"category_scores_gemma":[0.02434032,0.0008213079,0.002324809,0.001062926,0.0006074469,0.002722658,0.002380322,0.002651049,0.001680756],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001311137,"about_ca_system_score_gemma":0.002110266,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01465246,"about_ca_topic_score_gemma":0.02630757,"domain_scores_codex":[0.9966415,0.001987375,0.0001766923,0.0007758716,0.0002688156,0.0001497303],"domain_scores_gemma":[0.9696444,0.02760169,0.0005810566,0.0008884173,0.0009979384,0.0002865026],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0009666306,0.001196813,0.0307157,0.0007913307,0.0008284877,0.0008369461,0.003377064,0.3545096,0.01669937,0.009856369,0.02030843,0.5599132],"study_design_scores_gemma":[0.00002772398,0.00003980114,0.0007814449,0.00002023295,0.00005949443,0.00004920192,0.00008388539,0.9895926,0.001450619,0.006855968,0.001020745,0.00001832301],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07137357,0.0006821779,0.9151322,0.00157939,0.00007188573,0.0002704943,0.001166675,0.008612071,0.001111561],"genre_scores_gemma":[0.5572336,0.0004456808,0.4333064,0.001032012,0.0002226595,0.0004210271,0.004734848,0.0005615545,0.002042164],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01465246,"threshold_uncertainty_score":0.03735781,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01501122271640687,"score_gpt":0.2914003633171336,"score_spread":0.2763891406007267,"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."}}