{"id":"W4417132378","doi":"10.1109/bhi67747.2025.11269526","title":"Leveraging Image-to-Text Generators in Multimodal Vision Transformers for Inclusive Skin Cancer Diagnosis: A Comparative Study","year":2025,"lang":"","type":"article","venue":"","topic":"Cutaneous Melanoma Detection and Management","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Skin cancer; Transformer; Skin lesion; Fusion; Skin color; Feature (linguistics)","routes":{"ca_aff":true,"ca_fund":true,"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":[0.003143344,0.00176334,0.0007291554,0.001654635,0.0003297004,0.001208113,0.001409484,0.001316289,0.003451839],"category_scores_gemma":[0.008282651,0.0002727646,0.0008630061,0.0007499737,0.000514923,0.001713712,0.001444329,0.001253118,0.002335515],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008157538,"about_ca_system_score_gemma":0.0007766167,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005088788,"about_ca_topic_score_gemma":0.005678959,"domain_scores_codex":[0.9986626,0.0003733254,0.00009066635,0.0004724776,0.000265187,0.0001356724],"domain_scores_gemma":[0.9976024,0.001412983,0.0001082757,0.0003845419,0.0003434987,0.000148428],"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":[0.003291112,0.00172409,0.03582909,0.001172825,0.001037562,0.0004637268,0.0002375155,0.1200888,0.03265066,0.001137545,0.02716162,0.7752056],"study_design_scores_gemma":[0.0002239581,0.001682965,0.01651224,0.0001299767,0.0004111022,0.0008139823,0.0003554737,0.9157626,0.05185725,0.002605623,0.009572282,0.00007255436],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8235843,0.01017542,0.1157224,0.001603592,0.001030552,0.000815775,0.006354492,0.02581065,0.01490276],"genre_scores_gemma":[0.9348136,0.0008038551,0.04774654,0.0006494825,0.0001524395,0.0001118732,0.01166388,0.0003163057,0.003741965],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005088788,"threshold_uncertainty_score":0.0166238,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01859630719744177,"score_gpt":0.3645751030803192,"score_spread":0.3459787958828774,"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."}}