{"id":"W4413187029","doi":"10.1093/biomethods/bpaf051","title":"An explainable AI approach for mapping multivariate regional brain age and clinical severity patterns in Alzheimer’s disease","year":2025,"lang":"en","type":"article","venue":"Biology Methods and Protocols","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian Institutes of Health Research; National Institutes of Health; Genentech; IXICO; Servier; Eisai; Pfizer; Novartis Pharmaceuticals Corporation; H. Lundbeck A/S; Washington University in St. Louis; Biogen; BioClinica; Eli Lilly and Company; Bristol-Myers Squibb; U.S. Department of Defense; Meso Scale Diagnostics; National Institute on Aging; Alzheimer's Association; Mallinckrodt Institute of Radiology, School of Medicine, Washington University in St. Louis","keywords":"Multivariate statistics; Multivariate analysis; Disease; Psychology; Gerontology; Medicine; Internal medicine; Computer science; Machine learning","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005141242,0.0001518274,0.0003391144,0.0001182783,0.0001717419,0.0000762557,0.0003739441,0.0001913469,0.000001704871],"category_scores_gemma":[0.0006411819,0.0001301281,0.0000486192,0.0001555633,0.0001077584,0.0001483513,0.0002617766,0.0003355999,9.516177e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001408962,"about_ca_system_score_gemma":0.0001282275,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002692934,"about_ca_topic_score_gemma":0.00001267214,"domain_scores_codex":[0.995049,0.003458224,0.0004066036,0.0007385255,0.00004194006,0.0003057353],"domain_scores_gemma":[0.9978775,0.001389844,0.0001094239,0.0004132238,0.00004789861,0.0001621008],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001659495,0.0001362599,0.5243508,0.0003063775,0.00002730417,0.000008410205,0.0003035858,0.00001158451,0.00008364738,0.03234277,0.00008079113,0.4421825],"study_design_scores_gemma":[0.001273338,0.0003040207,0.6605499,0.0001037224,0.000007192537,0.000003478102,0.00003060334,0.2827159,0.00001276884,0.02995468,0.02483814,0.0002061834],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.00261885,0.0001132987,0.9639443,0.00370853,0.00005935565,0.0294513,0.00001036584,0.00005765838,0.00003637495],"genre_scores_gemma":[0.01233207,0.000005617328,0.9306179,0.003043335,0.00007049583,0.05387481,0.00002017751,0.000007349407,0.00002824432],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.4419763,"threshold_uncertainty_score":0.5306471,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1978260355470209,"score_gpt":0.5492460269426542,"score_spread":0.3514199913956333,"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."}}