{"id":"W3133501805","doi":"10.1186/s13195-021-00879-4","title":"Data analysis with Shapley values for automatic subject selection in Alzheimer’s disease data sets using interpretable machine learning","year":2021,"lang":"en","type":"article","venue":"Alzheimer s Research & Therapy","topic":"Dementia and Cognitive Impairment Research","field":"Medicine","cited_by":64,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute of Biomedical Imaging and Bioengineering; Canadian Institutes of Health Research; National Institutes of Health; Genentech; IXICO; H. Lundbeck A/S; Servier; Eisai; Northern California Institute for Research and Education; BioClinica; F. Hoffmann-La Roche; University of Southern California; Biogen; U.S. Department of Defense; Meso Scale Diagnostics; Alzheimer's Disease Neuroimaging Initiative; Novartis Pharmaceuticals Corporation; Pfizer; Eli Lilly and Company; Bristol-Myers Squibb; National Institute on Aging; Alzheimer's Association; Foundation for the National Institutes of Health","keywords":"Overfitting; Artificial intelligence; Machine learning; Feature selection; Random forest; Computer science; Test set; Cross-validation; Neuroimaging; Data mining; Medicine; Artificial neural network","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.03435674,0.001122611,0.001564471,0.003308074,0.0009155505,0.002677637,0.001449027,0.001304624,0.003443634],"category_scores_gemma":[0.09392504,0.0004386322,0.002292259,0.001908354,0.001657869,0.001873763,0.002212062,0.002471659,0.0007241523],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001004634,"about_ca_system_score_gemma":0.001768854,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001008625,"about_ca_topic_score_gemma":0.001408693,"domain_scores_codex":[0.9872377,0.007573713,0.001567564,0.001862374,0.0015493,0.0002093505],"domain_scores_gemma":[0.9454561,0.04068715,0.002794094,0.006997976,0.003490799,0.0005738357],"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":[0.003120277,0.0007258199,0.1016397,0.001098662,0.001970978,0.0007612152,0.001695153,0.1871078,0.009413403,0.02571261,0.01102393,0.6557304],"study_design_scores_gemma":[0.0001774289,0.000562522,0.01753729,0.0001640528,0.0001248896,0.0002032557,0.0002451199,0.9314046,0.005658379,0.0398335,0.00399799,0.00009084222],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06400028,0.0003451414,0.9295772,0.0003642875,0.0001082948,0.000814801,0.001386314,0.002678275,0.0007254478],"genre_scores_gemma":[0.3955623,0.0000978121,0.5995191,0.0001825622,0.0000624034,0.001453433,0.002455709,0.0002495846,0.0004171337],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.03435674,"threshold_uncertainty_score":0.181698,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2688370233607937,"score_gpt":0.4751454934411039,"score_spread":0.2063084700803102,"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."}}