{"id":"W3159629744","doi":"10.1109/jbhi.2021.3078127","title":"Estimation of Apnea-Hypopnea Index Using Deep Learning On 3-D Craniofacial Scans","year":2021,"lang":"en","type":"article","venue":"IEEE Journal of Biomedical and Health Informatics","topic":"Obstructive Sleep Apnea Research","field":"Medicine","cited_by":36,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute of Mental Health; National Institutes of Health; William Demant Fonden; Danmarks Tekniske Universitet; Center for Advanced Study in the Behavioral Sciences, Stanford University; Torben og Alice Frimodts Fond; Augustinus Fonden; Otto Mønsteds Fond; Knud Højgaards Fond; Sunovion; Vera og Carl Johan Michaelsens Legat; Stanford University; Reinholdt W. Jorck og Hustrus Fond; Jazz Pharmaceuticals; Klarman Family Foundation","keywords":"Craniofacial; Polysomnography; Obstructive sleep apnea; Artificial intelligence; Convolutional neural network; Sleep apnea; Hypopnea; Medicine; Computer science; Gold standard (test); Chin; Pattern recognition (psychology); Deep learning; Cephalometry; Random forest; Apnea; Orthodontics; Radiology; Cardiology; Internal medicine","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.0002736671,0.0008149379,0.0004076508,0.0008725625,0.0001773409,0.0003770575,0.0006350926,0.0005686759,0.0007883278],"category_scores_gemma":[0.001120976,0.0003319448,0.000716314,0.000415334,0.0002073945,0.0002537978,0.0004804477,0.000589336,0.0002574148],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005700701,"about_ca_system_score_gemma":0.0006225797,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02354806,"about_ca_topic_score_gemma":0.0273257,"domain_scores_codex":[0.999842,0.0000227875,0.00001045794,0.00005269741,0.00003876677,0.0000332444],"domain_scores_gemma":[0.9997671,0.00009991111,0.00003323189,0.00001799118,0.00006429956,0.00001742452],"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":[0.000312916,0.0002860395,0.05332861,0.00005997074,0.0001546622,0.0004709609,0.00007916259,0.7172523,0.01085389,0.000387347,0.00280927,0.2140048],"study_design_scores_gemma":[0.00000320272,0.00002287307,0.003448448,0.000004819669,0.000007820902,0.00003303208,0.00001083727,0.995304,0.0009120352,0.0001445334,0.0001039068,0.000004509836],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8041037,0.0007322445,0.1892679,0.000376208,0.00008451326,0.0001133972,0.001729542,0.001683114,0.001909363],"genre_scores_gemma":[0.9752085,0.0001576033,0.02222727,0.00007641464,0.00001830022,0.00005860093,0.00134009,0.00002923246,0.0008839785],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02354806,"threshold_uncertainty_score":0.04682201,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0480970144435272,"score_gpt":0.3711535155733243,"score_spread":0.3230565011297971,"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."}}