{"id":"W7117546582","doi":"10.2196/78838","title":"Evaluating GPT-4 Responses on Scars or Keloids for Patient Education: Large Language Model Evaluation Study","year":2025,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Readability; Unified Medical Language System; Scars; Language model; Reduction (mathematics); MEDLINE","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":[0.02899908,0.001106376,0.0009665867,0.001355783,0.0005022587,0.001585808,0.001101716,0.001165065,0.001948108],"category_scores_gemma":[0.1146157,0.0003028983,0.001536385,0.001111866,0.000790332,0.002019623,0.002097643,0.001581202,0.0008801373],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002194476,"about_ca_system_score_gemma":0.001844151,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003934519,"about_ca_topic_score_gemma":0.003651657,"domain_scores_codex":[0.9799478,0.01542919,0.00165904,0.001014189,0.001636431,0.0003133584],"domain_scores_gemma":[0.8426173,0.1319714,0.005129551,0.005099157,0.01320349,0.001978976],"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":[0.01746822,0.01825764,0.4079741,0.004943076,0.00223206,0.00146617,0.03152506,0.05141824,0.01488524,0.00148049,0.01636713,0.4319826],"study_design_scores_gemma":[0.004838798,0.0339271,0.3911587,0.0009829367,0.002849311,0.001698814,0.01473001,0.5059265,0.02383285,0.002583654,0.01692025,0.0005510832],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9900187,0.0001542276,0.006145641,0.0001831283,0.00002892957,0.001163081,0.0008379896,0.0002984063,0.001169781],"genre_scores_gemma":[0.9745342,0.0001966744,0.01857283,0.0002158493,0.00002711316,0.001945126,0.003729952,0.00008176791,0.0006965261],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02899908,"threshold_uncertainty_score":0.1533636,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2289841127160847,"score_gpt":0.5752512054612444,"score_spread":0.3462670927451597,"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."}}