{"id":"W7116777300","doi":"10.2196/83318","title":"Artificial Intelligence Models for Predicting Triage in Emergency Departments: Seven-Month Retrospective Comparative Study of Natural Language Processing, Large Language Model, and Joint Embedding Predictive Architectures","year":2025,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Data Visualization and Analytics","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Triage; Retrospective cohort study; Emergency department; Emergency response; MEDLINE","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":[0.0007759709,0.0001985408,0.000447761,0.000338203,0.0001331557,0.00007812167,0.0004504428,0.00009446149,0.000005205581],"category_scores_gemma":[0.0003538339,0.0001652813,0.0000518895,0.0006743291,0.000068046,0.0005783542,0.0003885232,0.000339,4.225795e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007325905,"about_ca_system_score_gemma":0.0001579312,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001305422,"about_ca_topic_score_gemma":0.000260463,"domain_scores_codex":[0.9976212,0.00007614141,0.001143151,0.0002380809,0.0006126654,0.0003088094],"domain_scores_gemma":[0.9989988,0.00009349985,0.0003787573,0.0002636981,0.0001576322,0.0001076211],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001965154,0.002155317,0.001767529,0.0008716522,0.0002048075,0.0000132067,0.8952168,0.05281219,0.00002845006,0.02229163,0.0004384736,0.02400343],"study_design_scores_gemma":[0.0005333964,0.0002266158,0.0003011592,0.0001706184,0.00001936273,0.000001375152,0.05788161,0.9369791,0.0001490158,0.003603831,0.000002181023,0.0001317208],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.351061,0.0001376084,0.647433,0.00003407218,0.00008079318,0.0008432603,0.0000588307,0.00006130729,0.000290107],"genre_scores_gemma":[0.993191,0.00001172045,0.006468314,0.0001058249,0.00002655448,0.000112575,0.0000403821,0.000006383338,0.00003722739],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8841669,"threshold_uncertainty_score":0.6739975,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0312730149266279,"score_gpt":0.3842273950826731,"score_spread":0.3529543801560452,"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."}}