{"id":"W4282962995","doi":"10.1038/s41597-022-01401-7","title":"A large, curated, open-source stroke neuroimaging dataset to improve lesion segmentation algorithms","year":2022,"lang":"en","type":"article","venue":"Scientific Data","topic":"Acute Ischemic Stroke Management","field":"Medicine","cited_by":149,"is_retracted":false,"has_abstract":true,"ca_institutions":"Research Institute for Aging; Heart and Stroke Foundation; University of Waterloo; Sunnybrook Health Science Centre; Ontario Brain Institute; University of Toronto; University of British Columbia","funders":"Eunice Kennedy Shriver National Institute of Child Health and Human Development; National Institute of Neurological Disorders and Stroke; Medical Research Council; National Institute of General Medical Sciences; Wellcome Trust; U.S. Department of Veterans Affairs","keywords":"Generalizability theory; Segmentation; Computer science; Artificial intelligence; Neuroimaging; Atlas (anatomy); Pattern recognition (psychology); Stroke (engine); Machine learning; Medicine; Mathematics; Statistics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[{"model":"gemma","categories":["open_science"],"domain":null,"study_design":"not_applicable","genre":"dataset","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":[],"domain":null,"study_design":"not_applicable","genre":"dataset","about_ca_system":false,"about_ca_topic":false,"confidence":"high","status":"direct model label, unvalidated"}],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["open_science","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001986715,0.0001699973,0.0002064057,0.0002688881,0.0007407909,0.0005445679,0.00234407,0.00001662714,0.0009140257],"category_scores_gemma":[0.0001650056,0.0001701369,0.00002516692,0.000814677,0.00006227687,0.0006860632,0.01677846,0.0002709288,0.000289814],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001608023,"about_ca_system_score_gemma":0.0001317039,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001371185,"about_ca_topic_score_gemma":0.00001358129,"domain_scores_codex":[0.9967853,0.00008566306,0.0003270994,0.001406676,0.000941292,0.0004539597],"domain_scores_gemma":[0.996333,0.00002922162,0.0001338777,0.003238953,0.00005859559,0.0002063541],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00005269705,0.0001919558,0.0001941771,0.00002299952,0.00003101597,0.00005929264,0.0001792082,0.00002812342,0.067858,0.000007394835,0.9059753,0.0253998],"study_design_scores_gemma":[0.00148405,0.0001068033,0.000158371,0.00001629821,0.0001007485,0.00003645288,0.003105585,0.01720556,0.003822917,0.000002379175,0.9737935,0.0001673213],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.1046675,0.0002834095,0.09574472,0.01752212,0.0114383,0.009417011,0.7517484,0.0005270113,0.008651554],"genre_scores_gemma":[0.0663624,0.00001262857,0.0368044,0.009346709,0.0003611394,0.0003702934,0.7949009,0.0001288231,0.0917127],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.08306114,"threshold_uncertainty_score":0.9999993,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06612173964472086,"score_gpt":0.3552075638803168,"score_spread":0.2890858242355959,"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."}}