{"id":"W4399115385","doi":"10.48550/arxiv.2405.16177","title":"Transformer Meets Gated Residual Networks To Enhance Photoplethysmogram Artifact Detection Informed by Mutual Information Neural Estimation","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Non-Invasive Vital Sign Monitoring","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Fonds de Recherche du Québec - Santé; Institut de Valorisation des Données; Centre hospitalier universitaire Sainte-Justine; Université de Montréal","keywords":"Residual; Photoplethysmogram; Artifact (error); Artificial neural network; Artificial intelligence; Computer science; Transformer; Mutual information; Pattern recognition (psychology); Engineering; Computer vision; Algorithm; Electrical engineering","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001503882,0.0004908909,0.0003272426,0.0004853954,0.0001219766,0.0002250605,0.0003385343,0.0004934459,0.0000261193],"category_scores_gemma":[0.00006312873,0.0006081666,0.0001644196,0.0008736713,0.00004746823,0.000958918,0.0001781222,0.001041162,0.0002598744],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007367085,"about_ca_system_score_gemma":0.00006433023,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001925545,"about_ca_topic_score_gemma":0.0002076178,"domain_scores_codex":[0.9983941,0.00004284733,0.0004309531,0.000448416,0.0001632073,0.0005205115],"domain_scores_gemma":[0.9990384,0.0001145133,0.0001067054,0.0003904098,0.0001175644,0.0002323839],"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":[0.00008096968,0.00001115727,0.00007459931,0.0003044211,0.0001094754,0.00001866422,0.0004641999,0.9688262,0.009904142,0.00004375598,0.0004144992,0.01974787],"study_design_scores_gemma":[0.0002697895,0.0001085088,0.0001661935,0.0002800374,0.0001483091,0.000004782044,0.0001177322,0.8329586,0.1637959,0.000560314,0.0008007461,0.0007889996],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6581766,0.00005335789,0.3381972,0.0000212492,0.001371081,0.0006771278,0.00005509851,0.0008381021,0.0006102006],"genre_scores_gemma":[0.9990405,0.0001081469,0.0002679516,0.00002967608,0.0001467945,0.00001294326,0.0002683497,0.000059644,0.00006602033],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3408639,"threshold_uncertainty_score":0.9996369,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01772741223515944,"score_gpt":0.1897013275848793,"score_spread":0.1719739153497199,"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."}}