{"id":"W4391252159","doi":"10.1101/2024.01.24.24301743","title":"Evaluating the Diagnostic and Treatment Recommendation Capabilities of GPT-4 Vision in Dermatology","year":2024,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"","keywords":"Dermatology; Medicine","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005572242,0.0001138851,0.0002624142,0.0001294927,0.00003653753,0.00001259648,0.00004038355,0.0001382517,0.00007057474],"category_scores_gemma":[0.001380755,0.00007387171,0.00004427474,0.00008782934,0.00009248352,0.00001421337,0.00009346273,0.0002827859,0.00001294889],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001621195,"about_ca_system_score_gemma":0.0002775087,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002490896,"about_ca_topic_score_gemma":0.0007971143,"domain_scores_codex":[0.9988446,0.0002027561,0.0004856407,0.0002474275,0.00009900695,0.0001206158],"domain_scores_gemma":[0.9977869,0.001757889,0.0001051129,0.0002272914,0.00008724171,0.00003553096],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.00008405152,0.0001905434,0.3716353,0.001662237,0.00004562028,0.00001016617,0.02430437,0.0001090276,0.0005002507,0.0003153053,0.0001138599,0.6010293],"study_design_scores_gemma":[0.0005017886,0.005542369,0.5664762,0.01051426,0.001343208,0.0003445259,0.03192362,0.1064771,0.06871203,0.2043035,0.003035045,0.0008263566],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.974579,0.001790872,0.00001180269,0.02191928,0.0007703521,0.0007205736,0.000005481212,0.00001555707,0.0001870336],"genre_scores_gemma":[0.9980295,0.001264192,0.0001569167,0.00008861446,0.0001283298,0.0002100204,0.00004653663,0.00001138891,0.00006452931],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6002029,"threshold_uncertainty_score":0.3765506,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2362795634093349,"score_gpt":0.506050948916747,"score_spread":0.2697713855074121,"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."}}