{"id":"W4417467340","doi":"10.1159/000550153","title":"When Jack of All Trades Is a Master of None: Comparing the Performance of GPT-4 Omni against Specialised Neural Networks in Identifying Malignant Dermatological Lesions from Smartphone Images and Structured Clinical Data","year":2025,"lang":"en","type":"article","venue":"Dermatology","topic":"Cutaneous Melanoma Detection and Management","field":"Medicine","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary; Queen's University; University of British Columbia; Public Health Ontario; University of Toronto","funders":"","keywords":"Skin lesion; Artificial neural network; Patient data; Dermatological diseases; Triage","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01225398,0.001849333,0.0009802138,0.001034403,0.0007427998,0.001904092,0.001583109,0.00214849,0.001540034],"category_scores_gemma":[0.03097088,0.0004622794,0.001188156,0.0004377069,0.001459753,0.003275717,0.002984532,0.001990772,0.0009045536],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001448962,"about_ca_system_score_gemma":0.001401374,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007392313,"about_ca_topic_score_gemma":0.007888066,"domain_scores_codex":[0.9962502,0.001605323,0.0002668622,0.001114576,0.0004604136,0.0003026575],"domain_scores_gemma":[0.9908956,0.005956252,0.0006703545,0.001146825,0.0008342572,0.0004966668],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0153842,0.00203448,0.1359667,0.0009649521,0.001773758,0.0009142087,0.001404194,0.2897469,0.01344022,0.002166895,0.02297562,0.5132279],"study_design_scores_gemma":[0.0003399631,0.002884915,0.02588871,0.0001885812,0.0003275505,0.000534175,0.0007069375,0.9466702,0.01286699,0.006081998,0.003362178,0.0001477938],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9610736,0.002120828,0.02562496,0.001825649,0.0005210852,0.0004411722,0.001136555,0.001947262,0.005308964],"genre_scores_gemma":[0.976835,0.0002534326,0.01827315,0.0009642412,0.0001324762,0.0001281432,0.001569713,0.0001347851,0.00170903],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01225398,"threshold_uncertainty_score":0.06480604,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08904350149265933,"score_gpt":0.3314629814763169,"score_spread":0.2424194799836576,"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."}}