{"id":"W4416239667","doi":"10.2196/66691","title":"Challenges and Solutions in Applying Large Language Models to Guideline-Based Management Planning and Automated Medical Coding in Health Care: Algorithm Development and Validation","year":2025,"lang":"en","type":"article","venue":"JMIR Biomedical Engineering","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Workflow; Generalizability theory; Coding (social sciences); Health informatics; Guideline; Decision support system; Precision and recall; Health care","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.08091557,0.001497839,0.001354052,0.002446332,0.001072831,0.005969285,0.003324806,0.002443698,0.001827388],"category_scores_gemma":[0.2241349,0.001098037,0.001626554,0.001789978,0.001947556,0.004499398,0.003635144,0.003369909,0.000700677],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003595746,"about_ca_system_score_gemma":0.009590224,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01693802,"about_ca_topic_score_gemma":0.01570463,"domain_scores_codex":[0.944759,0.04554583,0.002755638,0.00251398,0.003928747,0.000496651],"domain_scores_gemma":[0.7244505,0.2435018,0.005313283,0.01133092,0.01438548,0.001017995],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007668709,0.0007873575,0.03474981,0.00172485,0.0008325031,0.0003847261,0.002349279,0.3947338,0.004751857,0.02265381,0.01112922,0.5251359],"study_design_scores_gemma":[0.0001661586,0.0002372763,0.001793842,0.0005069691,0.00009804407,0.0001650095,0.0004999949,0.9643867,0.003761732,0.0233424,0.004979708,0.00006224326],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.09241661,0.003488626,0.8837078,0.01020322,0.0001937851,0.001282286,0.0009256316,0.005353253,0.002428735],"genre_scores_gemma":[0.2291813,0.0008383883,0.7663115,0.00113122,0.00006794051,0.0007741636,0.001114459,0.0003254123,0.0002555894],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.08091557,"threshold_uncertainty_score":0.4279276,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0250157991614873,"score_gpt":0.3356396931577726,"score_spread":0.3106238939962853,"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."}}