{"id":"W4406361577","doi":"10.1101/2025.01.13.25320425","title":"Improving Diagnostic Accuracy of Routine EEG for Epilepsy using Deep Learning","year":2025,"lang":"en","type":"preprint","venue":"medRxiv","topic":"EEG and Brain-Computer Interfaces","field":"Neuroscience","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal; Centre Hospitalier de l’Université de Montréal; Université de Montréal","funders":"Fonds de Recherche du Québec - Santé; Canadian Institutes of Health Research; Natural Sciences and Engineering Research Council of Canada; Fondation Brain Canada; Institut de Valorisation des Données; Eisai; Canada Research Chairs","keywords":"Epilepsy; Electroencephalography; Ictal; Cohort; Retrospective cohort study; Medicine; Receiver operating characteristic; Artificial intelligence; Deep learning; Convolutional neural network; Computer science; Machine learning; Surgery; Internal medicine; Psychiatry","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002948786,0.0007358732,0.0004332881,0.001452918,0.0001543844,0.0009163311,0.0005061723,0.0003757046,0.0008828662],"category_scores_gemma":[0.01305571,0.0002278038,0.0005650007,0.000529868,0.0003550855,0.0008381371,0.0008259148,0.0006784784,0.0002624114],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005444633,"about_ca_system_score_gemma":0.000567986,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002850789,"about_ca_topic_score_gemma":0.003685639,"domain_scores_codex":[0.9989447,0.0004081938,0.0001375798,0.0002584613,0.0001581299,0.00009302852],"domain_scores_gemma":[0.9937478,0.003871163,0.0009009587,0.0004234912,0.0008624488,0.000194216],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0009330488,0.0002661915,0.7015531,0.0002224618,0.0006213773,0.0003131896,0.0001250851,0.05104752,0.008461864,0.0004943968,0.002216669,0.2337451],"study_design_scores_gemma":[0.00006108666,0.0005948778,0.1267355,0.0001345918,0.0002539405,0.0008594109,0.0001073392,0.8571561,0.01082721,0.002206253,0.001012204,0.00005151582],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9322922,0.001506947,0.06290068,0.0006202807,0.00007386095,0.00004321929,0.0006771549,0.0005473979,0.00133811],"genre_scores_gemma":[0.9928681,0.0001504086,0.006482328,0.0000503949,0.00001884476,0.000007232411,0.0002964696,0.000009867299,0.0001163118],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.002948786,"threshold_uncertainty_score":0.0155949,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04182048302261464,"score_gpt":0.3160107785522456,"score_spread":0.274190295529631,"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."}}