{"id":"W4411799315","doi":"10.1109/tmi.2025.3583974","title":"Improving Robustness and Reliability in Medical Image Classification With Latent-Guided Diffusion and Nested-Ensembles","year":2025,"lang":"en","type":"article","venue":"IEEE Transactions on Medical Imaging","topic":"AI in cancer detection","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mila - Quebec Artificial Intelligence Institute; McGill University","funders":"Alliance de recherche numérique du Canada; Compute Canada; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Robustness (evolution); Computer science; Artificial intelligence; Reliability (semiconductor); Pattern recognition (psychology); Physics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008551892,0.0001699442,0.0002047106,0.0002781895,0.0002267432,0.0001288306,0.0002945261,0.0001361278,0.00001958036],"category_scores_gemma":[0.0001101281,0.0001430122,0.00002710465,0.000605431,0.0003358205,0.0005851205,0.00001555556,0.0006146258,0.000001162594],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001547317,"about_ca_system_score_gemma":0.0002602848,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003532904,"about_ca_topic_score_gemma":0.000209759,"domain_scores_codex":[0.9979163,0.0001603179,0.0003635912,0.0006301088,0.0006643941,0.0002653093],"domain_scores_gemma":[0.9989344,0.000333077,0.00006664977,0.0003632562,0.00008669691,0.0002158943],"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":[0.00005722214,0.0001915942,0.00346484,0.000171672,0.000009977851,0.00006917434,0.0002838282,0.0006752152,0.003042675,0.0001701272,0.00005649505,0.9918072],"study_design_scores_gemma":[0.001130516,0.00003129282,0.01108069,0.0004108454,0.00001630401,0.0001178872,0.0001002872,0.9844545,0.002276781,0.0001847148,0.00003819005,0.0001580581],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1758836,0.0000616706,0.8127651,0.01050274,0.0003656358,0.0001846365,6.452958e-7,0.000151698,0.00008435857],"genre_scores_gemma":[0.9893594,0.0001875471,0.009800549,0.0005233081,0.00002380843,0.00005551199,5.606155e-7,0.00001095909,0.00003839614],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9916492,"threshold_uncertainty_score":0.5831869,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01034731252026641,"score_gpt":0.2671102496100362,"score_spread":0.2567629370897698,"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."}}