{"id":"W4294975107","doi":"10.1109/embc48229.2022.9871372","title":"Data-Efficient Training of Pure Vision Transformers for the Task of Chest X-ray Abnormality Detection Using Knowledge Distillation","year":2022,"lang":"en","type":"article","venue":"2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)","topic":"COVID-19 diagnosis using AI","field":"Medicine","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"Health Sciences Centre; Sunnybrook Health Science Centre; York University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Transformer; Convolutional neural network; Abnormality; Artificial intelligence; Limiting; Machine learning; Training set; Pattern recognition (psychology); Voltage; Engineering; Electrical 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001904014,0.0001744119,0.0004244116,0.0001777505,0.0001191957,0.000003638263,0.0005441247,0.0001052282,0.00008618388],"category_scores_gemma":[0.0007485306,0.0001235921,0.0001872954,0.0005361663,0.0002545872,0.00006486125,0.0001630817,0.0004644702,1.677841e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002049625,"about_ca_system_score_gemma":0.0002338783,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003029835,"about_ca_topic_score_gemma":0.0001073163,"domain_scores_codex":[0.9983751,0.000096707,0.0006585974,0.0003054599,0.0003576744,0.0002065228],"domain_scores_gemma":[0.9980674,0.0008140312,0.0003516226,0.0003428778,0.000381052,0.00004299032],"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":[0.0005415418,0.0005016022,0.003266146,0.0008475503,0.0007050945,5.964929e-7,0.02287753,0.6004012,0.3611527,0.0004327256,0.003010529,0.006262735],"study_design_scores_gemma":[0.001747137,0.0003749256,0.0101299,0.0003826544,0.0002214564,0.00001557018,0.003612125,0.9610123,0.001526927,0.00003277413,0.02079486,0.0001493371],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8077307,0.00040517,0.1821601,0.005479152,0.002437005,0.0008144022,0.0009175985,0.00002804788,0.00002779544],"genre_scores_gemma":[0.9984592,0.00010399,0.0007435164,0.0001766352,0.0001826947,0.00004399161,0.0002376141,0.00001737903,0.0000350454],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3606111,"threshold_uncertainty_score":0.503994,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1130896754094058,"score_gpt":0.3805298655244511,"score_spread":0.2674401901150453,"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."}}