{"id":"W4319460119","doi":"10.1155/2023/3198066","title":"Using CNN Saliency Maps and EEG Modulation Spectra for Improved and More Interpretable Machine Learning‐Based Alzheimer’s Disease Diagnosis","year":2023,"lang":"en","type":"article","venue":"Computational Intelligence and Neuroscience","topic":"EEG and Brain-Computer Interfaces","field":"Neuroscience","cited_by":40,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Montreal Neurological Institute and Hospital; Institut National de la Recherche Scientifique; Université du Québec à Montréal","funders":"Natural Sciences and Engineering Research Council of Canada; Universidade Federal do ABC","keywords":"Spectrogram; Convolutional neural network; Electroencephalography; Artificial intelligence; Computer science; Pattern recognition (psychology); Benchmark (surveying); Modulation (music); Biomarker; Deep learning; Feature extraction; Machine learning; Neuroscience; Psychology; Biology","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002084596,0.0002122372,0.000174352,0.0002222741,0.00054878,0.000305342,0.0002178151,0.00003645918,0.000005352914],"category_scores_gemma":[0.0005538813,0.0001975646,0.00004350734,0.0005079614,0.0004880462,0.0004698663,0.0001835197,0.0001449425,0.00000244148],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001283741,"about_ca_system_score_gemma":0.0000468421,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001745605,"about_ca_topic_score_gemma":0.000001645198,"domain_scores_codex":[0.9982531,0.0000790238,0.0002737654,0.0008312664,0.0002304257,0.0003324315],"domain_scores_gemma":[0.998586,0.0009138725,0.000116866,0.0001188636,0.00006095016,0.0002034729],"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":[0.0001360555,0.0001264498,0.009229816,0.000108513,0.000004951717,0.00002369743,0.0007192444,0.8789042,0.07509732,0.003157041,0.00008781556,0.03240491],"study_design_scores_gemma":[0.0001249543,0.000182487,0.008974628,0.00004849063,0.00001695468,0.00001689725,0.00002912218,0.9633821,0.01966934,0.007060845,0.0002817439,0.0002124506],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.609943,0.0004053063,0.3866734,0.001631686,0.0004078631,0.0006186678,0.0001196462,0.000183918,0.0000165078],"genre_scores_gemma":[0.9959152,0.0002086399,0.002476556,0.001253259,0.0000307957,0.00003767708,0.00001457513,0.0000183737,0.00004487277],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3859722,"threshold_uncertainty_score":0.8056451,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1139947713310271,"score_gpt":0.3510165984430167,"score_spread":0.2370218271119895,"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."}}