{"id":"W3033107563","doi":"10.1016/j.nicl.2020.102290","title":"AD-NET: Age-adjust neural network for improved MCI to AD conversion prediction","year":2020,"lang":"en","type":"article","venue":"NeuroImage Clinical","topic":"Dementia and Cognitive Impairment Research","field":"Medicine","cited_by":61,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Johnson and Johnson Pharmaceutical Research and Development; National Institute of Biomedical Imaging and Bioengineering; Canadian Institutes of Health Research; Genentech; National Institutes of Health; H. Lundbeck A/S; GE Healthcare; Fujirebio US; Arizona State University; Janssen Alzheimer Immunotherapy Research And Development; Northern California Institute for Research and Education; Arizona Department of Health Services; U.S. Department of Defense; Alzheimer's Disease Neuroimaging Initiative; BrightFocus Foundation; University of Southern California; National Institute on Aging; Alzheimer's Association","keywords":"Neuroimaging; Transfer of learning; Deep learning; Artificial intelligence; Computer science; Machine learning; Artificial neural network; Cognition; Psychology; Neuroscience","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":[0.0005566533,0.0001942182,0.0004271882,0.00004575999,0.0001229497,0.00005086961,0.0001681213,0.0001327394,0.0003064706],"category_scores_gemma":[0.001839911,0.0001745178,0.0003198643,0.0002710733,0.000125342,0.0001088477,0.0001854306,0.0005599179,0.0002043353],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002220808,"about_ca_system_score_gemma":0.00009222286,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002833011,"about_ca_topic_score_gemma":0.000002102526,"domain_scores_codex":[0.9976123,0.0001816553,0.000611285,0.0007339057,0.0003238472,0.0005369975],"domain_scores_gemma":[0.9981555,0.0004581748,0.00008663157,0.0002881154,0.0002324966,0.0007790465],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"observational","study_design_scores_codex":[0.02004214,0.001433088,0.2009335,0.0005031132,0.000255702,0.0008328558,0.0002295144,0.00003407215,0.05376475,0.00001084389,0.5102796,0.2116808],"study_design_scores_gemma":[0.008980746,0.01323256,0.735415,0.00006192035,0.0003172909,0.00002068714,0.00007356006,0.03735003,0.0007088171,0.0000249834,0.2035852,0.000229181],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9446436,0.0001055029,0.005725974,0.04305705,0.001442346,0.003552982,0.0001187287,0.0002927796,0.00106109],"genre_scores_gemma":[0.9607219,0.0001405659,0.001966841,0.03352135,0.001741072,0.0001051518,0.0002177023,0.00006150325,0.00152398],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5344814,"threshold_uncertainty_score":0.7116631,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09844837713927006,"score_gpt":0.3955310145120361,"score_spread":0.2970826373727661,"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."}}