{"id":"W3156463266","doi":"10.1038/s41746-021-00440-5","title":"U-Sleep: resilient high-frequency sleep staging","year":2021,"lang":"en","type":"article","venue":"npj Digital Medicine","topic":"EEG and Brain-Computer Interfaces","field":"Neuroscience","cited_by":332,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Center for Advancing Translational Sciences; National Heart, Lung, and Blood Institute; Nvidia; Copenhagen Graduate School for Nanoscience and Nanotechnology; National Institutes of Health; Case Western Reserve University; Johns Hopkins University; Fundação para a Ciência e a Tecnologia; University of California, Davis; University of Minnesota; York University; University of Washington; American Sleep Medicine Foundation","keywords":"Polysomnography; Sleep (system call); Sleep Stages; Sleep medicine; Slow-wave sleep; Electroencephalography; Workflow; Convolutional neural network; Sleep study; Deep learning; Medicine; Artificial intelligence; Computer science; Sleep disorder; Insomnia; Psychiatry","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":[0.0008577569,0.0007371904,0.0003957651,0.000814246,0.0002691164,0.0004765116,0.0008458119,0.0006059703,0.001961203],"category_scores_gemma":[0.002745421,0.0002651148,0.0004978101,0.0004140727,0.0001789798,0.0006266434,0.001320508,0.0004393116,0.0007826464],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003180502,"about_ca_system_score_gemma":0.0005906392,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006450467,"about_ca_topic_score_gemma":0.01767856,"domain_scores_codex":[0.9996713,0.00007333739,0.00002962505,0.000125723,0.00005271937,0.00004727727],"domain_scores_gemma":[0.9995903,0.0001476983,0.00006219549,0.00007630757,0.00008072786,0.00004284939],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001960426,0.0003159874,0.1043692,0.0007432034,0.0005274109,0.0004971615,0.0006804361,0.03051644,0.03142592,0.001932044,0.05848048,0.7685513],"study_design_scores_gemma":[0.0003681781,0.001099363,0.118493,0.0003703604,0.0002941669,0.001465324,0.0003846621,0.7947777,0.03635518,0.009392293,0.03679838,0.0002013832],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5018414,0.005405755,0.4158502,0.001147126,0.0009305955,0.001005899,0.01609376,0.04829354,0.009431729],"genre_scores_gemma":[0.867253,0.0006686863,0.1160154,0.0006693093,0.000154844,0.0004649019,0.01046507,0.0006284923,0.003680164],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.006450467,"threshold_uncertainty_score":0.01282585,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02267243898740677,"score_gpt":0.2678749680301528,"score_spread":0.2452025290427461,"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."}}