{"id":"W4402167552","doi":"10.1200/jco.24.00326","title":"Enhancing Precision in Detecting Severe Immune-Related Adverse Events: Comparative Analysis of Large Language Models and International Classification of Disease Codes in Patient Records","year":2024,"lang":"en","type":"article","venue":"Journal of Clinical Oncology","topic":"Cancer Immunotherapy and Biomarkers","field":"Medicine","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Genentech; Foghorn Therapeutics; National Institute of Diabetes and Digestive and Kidney Diseases; National Heart, Lung, and Blood Institute; Mallinckrodt Pharmaceuticals; Massachusetts General Hospital; Eli Lilly and Company","keywords":"Medicine; Myocarditis; Pneumonitis; Internal medicine; Adverse effect","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.03951484,0.001177227,0.001705507,0.0050258,0.0005654584,0.00299899,0.001137992,0.001307087,0.0007022487],"category_scores_gemma":[0.117056,0.0004547134,0.002644943,0.00242169,0.0008294149,0.002271331,0.001836478,0.001381676,0.0003891319],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001097878,"about_ca_system_score_gemma":0.001331021,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005462447,"about_ca_topic_score_gemma":0.005597875,"domain_scores_codex":[0.9768052,0.0158479,0.001998892,0.003219897,0.001697159,0.000430866],"domain_scores_gemma":[0.8102487,0.1701062,0.009182296,0.005893976,0.003561684,0.001007173],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.003404377,0.0004279365,0.8671694,0.0006802158,0.003355469,0.000251605,0.0007714764,0.02597851,0.001081172,0.0003995344,0.002641529,0.09383881],"study_design_scores_gemma":[0.0005029191,0.002147149,0.4697205,0.0004996897,0.002227373,0.0009577564,0.001241802,0.5124583,0.002220022,0.005295093,0.002508722,0.0002207336],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9595976,0.003995553,0.02791303,0.001662089,0.0001624918,0.0002870189,0.00430385,0.0007281935,0.001350199],"genre_scores_gemma":[0.9820799,0.0004482545,0.01337321,0.0002203875,0.0001422378,0.00009556171,0.003456937,0.00003990373,0.0001436903],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9604852,"threshold_uncertainty_score":0.2089769,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07668197061795526,"score_gpt":0.4631874951858104,"score_spread":0.3865055245678551,"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."}}