{"id":"W4392607662","doi":"10.1109/jbhi.2024.3368500","title":"MAD-Former: A Traceable Interpretability Model for Alzheimer's Disease Recognition Based on Multi-Patch Attention","year":2024,"lang":"en","type":"article","venue":"IEEE Journal of Biomedical and Health Informatics","topic":"Dementia and Cognitive Impairment Research","field":"Medicine","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute on Aging; National Institute of Biomedical Imaging and Bioengineering; Canadian Institutes of Health Research; Genentech; National Institutes of Health; H. Lundbeck A/S; Servier; National Natural Science Foundation of China; Eisai; Takeda Pharmaceutical Company; IXICO; Northern California Institute for Research and Education; Pfizer; Biogen; BioClinica; AbbVie; Roche; University of Southern California; Novartis Pharmaceuticals Corporation; U.S. Department of Defense; Eli Lilly and Company; Bristol-Myers Squibb; GE Healthcare; Alzheimer's Disease Neuroimaging Initiative; Meso Scale Diagnostics; Alzheimer's Drug Discovery Foundation; Merck; Alzheimer's Association; National Science Foundation","keywords":"Interpretability; Computer science; Artificial intelligence; Convolutional neural network; Feature extraction; Pattern recognition (psychology); Deep learning; Feature (linguistics); 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009344359,0.001059808,0.001126076,0.0009088421,0.0003011298,0.001041942,0.002346602,0.001187949,0.002000332],"category_scores_gemma":[0.002310096,0.0003817769,0.001564146,0.0004285163,0.0005806695,0.001369869,0.001234212,0.00169879,0.0003868477],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001419207,"about_ca_system_score_gemma":0.0007310584,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009257944,"about_ca_topic_score_gemma":0.00846047,"domain_scores_codex":[0.9995592,0.00007210555,0.00002609048,0.0001903582,0.00008964742,0.00006262506],"domain_scores_gemma":[0.9993881,0.0002794739,0.0000831425,0.00007874567,0.0001280074,0.00004259377],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005239616,0.0002872765,0.004809034,0.0001533545,0.0003697427,0.0003746162,0.0002099544,0.5418759,0.01998059,0.01295336,0.00534822,0.413114],"study_design_scores_gemma":[0.000008414345,0.00004667417,0.0004762679,0.000004433498,0.00002572227,0.0000423025,0.000004502877,0.9938611,0.001172005,0.0039572,0.0003946875,0.000006719397],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0412828,0.000801837,0.9532382,0.0005485203,0.00009287909,0.00007071127,0.0002960473,0.002055365,0.001613594],"genre_scores_gemma":[0.8841045,0.000467154,0.1083006,0.0005553424,0.000165913,0.0001493937,0.0008230505,0.0001928542,0.00524117],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009257944,"threshold_uncertainty_score":0.01840812,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09961327850566637,"score_gpt":0.403747000303852,"score_spread":0.3041337217981856,"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."}}