{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001169639,0.0001310032,0.0001965162,0.0006574128,0.00008293303,0.00003991036,0.0001596063,0.00009364165,6.13472e-7],"category_scores_gemma":[0.00006792416,0.0001373816,0.000007396317,0.0004319813,0.0000146639,0.00008922687,0.0003650053,0.0002413908,2.524278e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001724119,"about_ca_system_score_gemma":0.0001159368,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006853345,"about_ca_topic_score_gemma":0.00003375812,"domain_scores_codex":[0.9984421,0.00005795357,0.0004295602,0.0003798341,0.0003082964,0.0003822955],"domain_scores_gemma":[0.9994943,0.0001040282,0.00003369446,0.0001299111,0.00001564534,0.0002224438],"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":[0.000003263528,0.0000437979,0.0002795071,0.001724854,0.00001217625,0.0001026583,0.01542926,0.02462653,0.00001492591,0.007815755,0.00006240933,0.9498848],"study_design_scores_gemma":[0.0005893449,0.00002029859,0.005448715,0.001288864,7.799343e-7,0.000004797697,0.0006361137,0.9900107,0.00000600539,0.00001474905,0.001863844,0.0001157775],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01271687,0.006074817,0.9743764,0.00583084,0.00009920872,0.0005213043,0.000001706693,0.0003409107,0.00003793416],"genre_scores_gemma":[0.7047725,0.0001224699,0.2942789,0.0005266047,0.00002066304,0.0002426947,0.00002214741,0.0000108142,0.000003235175],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9653842,"threshold_uncertainty_score":0.560226,"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."}}