{"id":"W3169282546","doi":"10.1016/j.ymeth.2021.06.002","title":"Learning from imbalanced COVID-19 chest X-ray (CXR) medical imaging data","year":2021,"lang":"en","type":"article","venue":"Methods","topic":"COVID-19 diagnosis using AI","field":"Medicine","cited_by":7,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"King Mongkut's University of Technology Thonburi; University of Toronto","keywords":"Coronavirus disease 2019 (COVID-19); Computer science; 2019-20 coronavirus outbreak; Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2); Software deployment; Pandemic; Task (project management); Medical imaging; Data science; Set (abstract data type); Artificial intelligence; Medicine; Pathology","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":["metaresearch","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.004000682,0.0002411237,0.0006199547,0.0001099942,0.0001851033,0.00007613827,0.0005359405,0.0001730612,0.004791428],"category_scores_gemma":[0.05538999,0.0002334274,0.0001091274,0.0005094362,0.000159351,0.0001376102,0.0008876015,0.0008970483,0.0001022343],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002424803,"about_ca_system_score_gemma":0.002015203,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000956643,"about_ca_topic_score_gemma":0.0000566581,"domain_scores_codex":[0.9959081,0.001443238,0.0004504375,0.001019346,0.0007357339,0.0004431304],"domain_scores_gemma":[0.9916809,0.005501095,0.0001358628,0.001725468,0.0001283473,0.000828382],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001641786,0.0005901328,0.1406758,0.0005434971,0.0003812784,0.005172392,0.002272341,0.00033268,0.07528977,0.00007580323,0.1423375,0.6321647],"study_design_scores_gemma":[0.002080675,0.00002846619,0.01667026,0.0002573266,0.0002455093,0.0001029275,0.000492579,0.04387968,0.004668299,0.000219591,0.9310783,0.0002763798],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0105457,0.004378323,0.7127148,0.2690722,0.001252349,0.0002727146,0.00004527279,0.0006438266,0.00107482],"genre_scores_gemma":[0.03620563,0.0005654212,0.7884969,0.1702795,0.001581402,0.00003429106,0.00139737,0.00012401,0.001315582],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7887409,"threshold_uncertainty_score":0.9961183,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1281199592434513,"score_gpt":0.4820925101165937,"score_spread":0.3539725508731424,"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."}}