{"id":"W4407637147","doi":"10.2196/67551","title":"Evaluating the Diagnostic Accuracy of ChatGPT-4 Omni and ChatGPT-4 Turbo in Identifying Melanoma: Comparative Study","year":2025,"lang":"en","type":"article","venue":"JMIR Dermatology","topic":"AI in cancer detection","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Preprint; Melanoma; Computer science; Medicine; Cancer research; World Wide Web","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":[0.0004563301,0.0001526426,0.0003710403,0.0002592515,0.0001585162,0.00007100468,0.0006446618,0.00007434237,0.000005049641],"category_scores_gemma":[0.0003777629,0.0001232945,0.00003398315,0.000748347,0.0001348771,0.0003438403,0.0005258769,0.0002652352,0.000007743709],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006500799,"about_ca_system_score_gemma":0.0001020614,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001168093,"about_ca_topic_score_gemma":0.0002948255,"domain_scores_codex":[0.9980946,0.0005680312,0.0004662324,0.0004097999,0.0002088969,0.0002524106],"domain_scores_gemma":[0.9962909,0.002832206,0.0002452102,0.0005161641,0.00008938919,0.00002616628],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001834093,0.001208524,0.6913726,0.0006663305,0.0003948198,0.0002512082,0.1495926,0.00118661,0.003189245,0.02573497,0.001586551,0.1246332],"study_design_scores_gemma":[0.00310458,0.0005074952,0.8446563,0.0002765127,0.00005456067,0.000287084,0.005392526,0.1233492,0.009433666,0.01226462,0.0003018506,0.0003715222],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9869955,0.001094293,0.008261739,0.001527748,0.0005190507,0.001173589,7.349381e-7,0.00005249468,0.0003748358],"genre_scores_gemma":[0.9986954,0.00002667843,0.0006553495,0.0001742383,0.00001667194,0.0004047809,5.208366e-7,0.000004854466,0.00002148522],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1532838,"threshold_uncertainty_score":0.5027803,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06754269392955682,"score_gpt":0.4176163842670595,"score_spread":0.3500736903375027,"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."}}