{"id":"W4416873224","doi":"10.1109/access.2025.3639184","title":"Breast Cancer Diagnosis With Explainable Artificial Intelligence (XAI): Uncovering Strengths and Biases","year":2025,"lang":"en","type":"article","venue":"IEEE Access","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of New Brunswick","funders":"","keywords":"Breast cancer; Transparency (behavior); Clinical Practice; Convolutional neural network; Artificial neural network; Mammography","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":[0.01227615,0.0008284505,0.000835444,0.002383249,0.0004537913,0.003817336,0.001586498,0.001272386,0.001722404],"category_scores_gemma":[0.07797866,0.0004010591,0.0007921393,0.001647478,0.001848042,0.004356469,0.003237166,0.002688953,0.0002761337],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001809291,"about_ca_system_score_gemma":0.001336516,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002360078,"about_ca_topic_score_gemma":0.001906907,"domain_scores_codex":[0.9938979,0.003451809,0.0003592969,0.0007863533,0.001298283,0.0002063641],"domain_scores_gemma":[0.9521528,0.03780926,0.00348131,0.003971571,0.002252729,0.0003323338],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003563481,0.0001723641,0.0894099,0.001588408,0.001030318,0.0004060499,0.001507081,0.1141723,0.00194053,0.2401493,0.007831629,0.5414356],"study_design_scores_gemma":[0.00003563334,0.0001335221,0.01184192,0.0005120859,0.0001772356,0.00034535,0.0003000011,0.4715168,0.002144737,0.5047898,0.008136942,0.00006602713],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1252053,0.01183015,0.8210421,0.02593416,0.0002876026,0.0002616255,0.001403836,0.001242795,0.01279241],"genre_scores_gemma":[0.8677008,0.003627444,0.1251309,0.001428863,0.0003042088,0.0001765754,0.0006268948,0.00009332919,0.0009109643],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01227615,"threshold_uncertainty_score":0.06492323,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04088765908412852,"score_gpt":0.3445365481135397,"score_spread":0.3036488890294112,"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."}}