{"id":"W4414811871","doi":"10.1701/4573.45774","title":"The path to trustworthy medical AI: the evolving role of explainability","year":2025,"lang":"en","type":"article","venue":"Recenti Progressi in Medicina","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Trustworthiness; Meaning (existential); Path (computing); Tracing; Connection (principal bundle); Expert system","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":[0.003135427,0.0001180554,0.0002726332,0.0001072399,0.0002335135,0.00001919418,0.0003107279,0.0001404229,0.0001612811],"category_scores_gemma":[0.005637427,0.00006239275,0.00006546362,0.0008482894,0.0003468519,0.00004711849,0.00008421848,0.0005323487,0.00001177514],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001706961,"about_ca_system_score_gemma":0.0009011609,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00051045,"about_ca_topic_score_gemma":0.0004231177,"domain_scores_codex":[0.9977382,0.0002326222,0.0007104906,0.0002581676,0.0006770805,0.0003834303],"domain_scores_gemma":[0.9979782,0.0008277249,0.0001045792,0.0005503151,0.0003628995,0.0001762833],"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":[0.0001355149,0.0001683012,0.2995179,0.0001497652,0.00001513265,0.00000559063,0.002569561,7.825022e-7,0.00003111983,0.001224517,0.004127474,0.6920543],"study_design_scores_gemma":[0.0006614925,0.001296051,0.4666981,0.01033594,0.0002149246,0.00005189968,0.07679818,0.008540892,0.01644892,0.03493131,0.3836158,0.0004064792],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8290937,0.01374042,0.0001951172,0.1489577,0.001547736,0.001739678,0.000001087599,0.00004313772,0.004681407],"genre_scores_gemma":[0.9965203,0.0008709209,0.0001111132,0.001823622,0.0002494112,0.0002058875,0.00000481131,0.000008841456,0.0002050385],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6916478,"threshold_uncertainty_score":0.6748934,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03226275652349358,"score_gpt":0.4308648204999115,"score_spread":0.398602063976418,"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."}}