{"id":"W4391220613","doi":"10.2196/53378","title":"A Machine Learning Approach with Human-AI Collaboration for Automated Classification of Patient Safety Event Reports: Algorithm Development and Validation Study","year":2024,"lang":"en","type":"article","venue":"JMIR Human Factors","topic":"Patient Safety and Medication Errors","field":"Health Professions","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Agency for Healthcare Research and Quality","keywords":"Computer science; Artificial intelligence; Machine learning; Classifier (UML); Confusion matrix; Confusion; Natural language processing","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.007313759,0.001135203,0.001038043,0.001489201,0.0008366099,0.001404498,0.002061055,0.002255463,0.002047814],"category_scores_gemma":[0.01627878,0.0003675878,0.0008509827,0.001250673,0.000483146,0.001295692,0.001431594,0.001977269,0.000741537],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001476803,"about_ca_system_score_gemma":0.00239087,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008935682,"about_ca_topic_score_gemma":0.005532947,"domain_scores_codex":[0.9974146,0.001266045,0.0002424356,0.0004922847,0.0004003203,0.0001843273],"domain_scores_gemma":[0.9882819,0.008512756,0.0003675274,0.0005854303,0.002084043,0.0001682516],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005336014,0.00111471,0.01901495,0.0001854693,0.0003039833,0.0002181202,0.0003433156,0.3214743,0.002961074,0.001829408,0.002813196,0.6492078],"study_design_scores_gemma":[0.00002567268,0.00006701115,0.0006925638,0.00001012079,0.0000167793,0.00002243255,0.00004859967,0.9976706,0.0006993735,0.0005122023,0.0002303951,0.000004311027],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2434857,0.001620432,0.745028,0.001042756,0.0001122752,0.0009917028,0.0002327561,0.004609889,0.002876516],"genre_scores_gemma":[0.6129534,0.0002969226,0.3834041,0.0002689882,0.00004534036,0.0008866005,0.0006503949,0.00007081858,0.001423367],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9926862,"threshold_uncertainty_score":0.0386793,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06328858342681644,"score_gpt":0.4205311001836615,"score_spread":0.3572425167568451,"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."}}