{"id":"W4415677754","doi":"10.2196/76384","title":"Detecting Laterality Errors in Combined Radiographic Studies by Enhancing the Traditional Approach With GPT-4o: Algorithm Development and Multisite Internal Validation","year":2025,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Radiology practices and education","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Laterality; Radiography; Quality assurance; Pattern recognition (psychology); Quality (philosophy)","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":[0.01184501,0.002310291,0.001927246,0.004382461,0.0009614941,0.003847013,0.004328718,0.002568201,0.001686247],"category_scores_gemma":[0.03271122,0.0009185374,0.002347146,0.002347696,0.001114494,0.002257213,0.003785089,0.003279218,0.001490682],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001260369,"about_ca_system_score_gemma":0.003246912,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01020893,"about_ca_topic_score_gemma":0.01240848,"domain_scores_codex":[0.9940163,0.001504849,0.0007605855,0.001949716,0.001443063,0.0003254588],"domain_scores_gemma":[0.9849852,0.007271612,0.001221132,0.002726254,0.003359153,0.0004366256],"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":[0.0005578778,0.0005251135,0.0744023,0.0005266767,0.0009184122,0.0007667755,0.0004234318,0.1542891,0.008351914,0.001355401,0.01926899,0.7386139],"study_design_scores_gemma":[0.00008639788,0.0001732711,0.007365103,0.0001433357,0.0002034883,0.000673811,0.0001687621,0.9776859,0.005307081,0.003918994,0.004233482,0.00004036548],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1028914,0.00305128,0.8634405,0.001468425,0.0004795384,0.001047756,0.002853634,0.02205637,0.002711076],"genre_scores_gemma":[0.3835039,0.0006806625,0.6022567,0.0008159032,0.0002338589,0.0006005162,0.009649522,0.0008297625,0.001429193],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01184501,"threshold_uncertainty_score":0.06264311,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09402885159616922,"score_gpt":0.4206768109893587,"score_spread":0.3266479593931895,"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."}}