{"id":"W3126652400","doi":"10.1038/s41598-021-82760-w","title":"Assessing robustness of carotid artery CT angiography radiomics in the identification of culprit lesions in cerebrovascular events","year":2021,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":50,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"NIHR Imperial Biomedical Research Centre; Horizon 2020; Medical Research Council; NIHR Cambridge Biomedical Research Centre; European Commission; School of Clinical Medicine, University of Cambridge; Wellcome Trust; Royal College of Surgeons of England; Dunhill Medical Trust; Department of Health and Social Care; Leverhulme Trust; National Institute for Health and Care Research; Mark Foundation For Cancer Research; Higher Education Funding Council for England; Alan Turing Institute; University of Cambridge; Cancer Research UK; British Heart Foundation; Engineering and Physical Sciences Research Council; AstraZeneca","keywords":"Radiomics; Medicine; Robustness (evolution); Radiology; Artificial intelligence; Culprit; Carotid artery disease; Segmentation; Computed tomography angiography; Feature extraction; Angiography; Feature (linguistics); Pattern recognition (psychology); Computer science; Carotid endarterectomy; Internal medicine","routes":{"ca_aff":true,"ca_fund":false,"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":[0.004622748,0.000102878,0.0003428868,0.0005166379,0.00007989448,0.0000750897,0.0001332161,0.00004113921,0.00002138261],"category_scores_gemma":[0.000826476,0.00008183644,0.0002951124,0.001818181,0.0002281425,0.0001684391,0.00004368151,0.0002718104,5.3475e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004076354,"about_ca_system_score_gemma":0.0002669611,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007338676,"about_ca_topic_score_gemma":0.00002090613,"domain_scores_codex":[0.9974079,0.0002140965,0.001038217,0.0004532805,0.0006758515,0.0002106303],"domain_scores_gemma":[0.9983634,0.00008289675,0.0004419988,0.0008683262,0.0001861026,0.00005730963],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.000005286485,0.0007146437,0.8474535,0.000367716,0.00008706271,0.0008883691,0.001136895,0.005879879,0.1374751,0.00005793678,0.0008110389,0.005122646],"study_design_scores_gemma":[0.0009220004,0.0000189234,0.9456492,0.001100284,0.0001919788,0.00262908,0.002254888,0.01210004,0.0327642,0.001250012,0.0009215996,0.0001978041],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9945443,0.0004525128,0.002400507,0.0005134747,0.001606841,0.0002732999,8.359908e-7,0.000009116532,0.0001991288],"genre_scores_gemma":[0.9983549,0.00002519835,0.001361084,0.00002514003,0.00003405753,0.00001123623,0.00007413862,0.00001191786,0.0001023965],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1047109,"threshold_uncertainty_score":0.3337193,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01756022279446152,"score_gpt":0.3003266172250981,"score_spread":0.2827663944306366,"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."}}