{"id":"W4288053873","doi":"10.1186/s40708-022-00165-5","title":"A robust framework to investigate the reliability and stability of explainable artificial intelligence markers of Mild Cognitive Impairment and Alzheimer’s Disease","year":2022,"lang":"en","type":"article","venue":"Brain Informatics","topic":"Dementia and Cognitive Impairment Research","field":"Medicine","cited_by":61,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute on Aging; European Regional Development Fund; National Institute of Biomedical Imaging and Bioengineering; Canadian Institutes of Health Research; National Institutes of Health; Genentech; IXICO; H. Lundbeck A/S; Servier; Eisai; Ministero dell’Istruzione, dell’Università e della Ricerca; Northern California Institute for Research and Education; Pfizer; Biogen; BioClinica; Foundation for the National Institutes of Health; Regione Puglia; F. Hoffmann-La Roche; University of Southern California; Eli Lilly and Company; U.S. Department of Defense; Meso Scale Diagnostics; Alzheimer's Disease Neuroimaging Initiative; Novartis Pharmaceuticals Corporation; Bristol-Myers Squibb; Alzheimer's Association","keywords":"Dementia; Neurocognitive; Neuropsychology; Disease; Cognition; Reliability (semiconductor); Psychology; Multivariate statistics; Alzheimer's disease; Multivariate analysis; Neuropsychological test; Cognitive impairment; Clinical psychology; Artificial intelligence; Medicine; Machine learning; Computer science; Psychiatry; Pathology","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":[0.00211163,0.0001295461,0.0002460288,0.0001052916,0.0001982425,0.0000216868,0.0001104457,0.00002906423,0.0002355759],"category_scores_gemma":[0.001850874,0.00009873904,0.00005569844,0.0003843871,0.0005658999,0.0001152711,0.0004771548,0.0002755783,0.000001684709],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004451019,"about_ca_system_score_gemma":0.0002393048,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004545401,"about_ca_topic_score_gemma":0.000002677981,"domain_scores_codex":[0.9982111,0.0001951225,0.0006113563,0.0001476014,0.0005768068,0.000257997],"domain_scores_gemma":[0.9981131,0.0009211481,0.000163646,0.0002744648,0.0002381158,0.0002895005],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.02516881,0.00366505,0.7603486,0.007824678,0.0010421,0.00003781739,0.1059515,0.001576513,0.000302091,0.008196791,0.003792566,0.0820935],"study_design_scores_gemma":[0.001839824,0.008637896,0.6325856,0.001244621,0.0009824606,0.00003169583,0.2680921,0.04413487,0.01347953,0.027521,0.0008125166,0.0006378742],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9918794,0.0001444815,0.002720218,0.002899041,0.00004048794,0.001865395,0.0001385051,0.00001207577,0.0003004147],"genre_scores_gemma":[0.9964305,0.00002867337,0.0022115,0.001145493,0.00001007083,0.0001319897,0.00001922355,0.000007613306,0.0000149392],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1621405,"threshold_uncertainty_score":0.4026461,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05881664022996328,"score_gpt":0.315065517541411,"score_spread":0.2562488773114477,"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."}}