{"id":"W4388975157","doi":"10.3389/fnins.2023.1302132","title":"Machine-learning assisted swallowing assessment: a deep learning-based quality improvement tool to screen for post-stroke dysphagia","year":2023,"lang":"en","type":"article","venue":"Frontiers in Neuroscience","topic":"Dysphagia Assessment and Management","field":"Health Professions","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"Centre for Social Innovation; University of Toronto; Toronto Metropolitan University; Health Sciences Centre; University of Waterloo; Sunnybrook Health Science Centre","funders":"Temerty Faculty of Medicine, University of Toronto; Sunnybrook Research Institute","keywords":"Dysphagia; Swallowing; Stroke (engine); Physical medicine and rehabilitation; Medicine; Deep learning; Physical therapy; Computer science; Artificial intelligence; Surgery; Engineering","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":[0.002397351,0.0007170451,0.0006828695,0.0006422437,0.0001553739,0.0006758947,0.0007490037,0.0007132857,0.001136312],"category_scores_gemma":[0.004557522,0.0002187831,0.0006235006,0.0003630426,0.0002181842,0.000483202,0.0007156085,0.0006517852,0.0003476519],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007665572,"about_ca_system_score_gemma":0.0008831001,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003231463,"about_ca_topic_score_gemma":0.004295876,"domain_scores_codex":[0.9994306,0.0002229662,0.0000396936,0.000113844,0.0001489564,0.00004401838],"domain_scores_gemma":[0.9990091,0.0004535349,0.0001450566,0.00007898998,0.0002527472,0.0000605561],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001908144,0.001405436,0.09101814,0.0004897177,0.0008411599,0.0003657053,0.0001311676,0.149692,0.01569142,0.001079589,0.01020033,0.7271772],"study_design_scores_gemma":[0.0001212594,0.0008957927,0.02168633,0.0000775332,0.000159393,0.0002013035,0.00001969817,0.9683679,0.005847975,0.001030956,0.001555033,0.00003672054],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.522468,0.003780991,0.4571038,0.002513413,0.0002362399,0.0007353675,0.002701813,0.006168457,0.004291867],"genre_scores_gemma":[0.9069895,0.0005840444,0.08930219,0.0004514975,0.00005918395,0.0003202907,0.001097702,0.00004045868,0.001155176],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.003231463,"threshold_uncertainty_score":0.01267862,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04167400148496919,"score_gpt":0.3956762834874314,"score_spread":0.3540022820024622,"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."}}