{"id":"W2952735543","doi":"10.1109/tmi.2019.2905770","title":"Standardized Assessment of Automatic Segmentation of White Matter Hyperintensities and Results of the WMH Segmentation Challenge","year":2019,"lang":"en","type":"article","venue":"IEEE Transactions on Medical Imaging","topic":"Medical Image Segmentation Techniques","field":"Computer Science","cited_by":301,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph; University of British Columbia; Montreal Neurological Institute and Hospital; Toronto Metropolitan University; University of Calgary; McGill University","funders":"National Institute of Biomedical Imaging and Bioengineering; National Institute of Allergy and Infectious Diseases; National Institutes of Health; Hotchkiss Brain Institute, University of Calgary; Universitair Medisch Centrum Utrecht; Institute for Basic Science; Daegu Gyeongbuk Institute of Science and Technology; University College London Hospitals NHS Foundation Trust; Télécom Paris; Multiple Sclerosis Society; Ministry of Advanced Education; Schweizerische Multiple Sklerose Gesellschaft; Ministerio de Ciencia y Tecnología; University of British Columbia; Huazhong University of Science and Technology; Natural Sciences and Engineering Research Council of Canada; Ministerio de Economía y Competitividad; Leids Universitair Medisch Centrum; National Natural Science Foundation of China; National Research Foundation of Korea; Ministerio de Educación, Cultura y Deporte; Sun Yat-sen University; Coordenação de Aperfeiçoamento de Pessoal de Nível Superior; ZonMw; Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung; European Regional Development Fund; King's College London; National Research Foundation; Alzheimer Society; National Institute for Health and Care Research; Nederlandse Organisatie voor Wetenschappelijk Onderzoek; Technische Universität München; Canadian Institutes of Health Research; Alzheimer's Society; Inselspital, Universitätsspital Bern; Hotchkiss Brain Institute; Skolkovo Institute of Science and Technology; Ministry of Advanced Education and Skills Development; Brigham and Women's Hospital; National University Health System; Nvidia; Ministry of Education; University of Bern; Université Paris-Saclay; Universiteit Utrecht; Universität Basel; University of Dundee; McGill University; Universitat Politècnica de Catalunya; Sungkyunkwan University; Universitat de Girona; University College London; Vrije Universiteit Amsterdam; National Science Foundation","keywords":"Segmentation; Hyperintensity; Artificial intelligence; Robustness (evolution); Scanner; Computer science; Fluid-attenuated inversion recovery; Percentile; Hausdorff distance; Pattern recognition (psychology); Image segmentation; Sørensen–Dice coefficient; Computer vision; Mathematics; Magnetic resonance imaging; Statistics; Medicine; Radiology","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":[0.0007904036,0.0001197168,0.000271735,0.0001548526,0.00006606469,0.00001772378,0.0003034855,0.00004525448,0.0002365614],"category_scores_gemma":[0.00002682972,0.00008982647,0.00008129203,0.0002270663,0.0002486816,0.0003469814,0.00001650729,0.0001927254,0.00000210486],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005443872,"about_ca_system_score_gemma":0.0001160323,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004122214,"about_ca_topic_score_gemma":0.000004156789,"domain_scores_codex":[0.9976552,0.0002064085,0.0007215768,0.0002356018,0.001048728,0.0001325013],"domain_scores_gemma":[0.9987863,0.0002399232,0.000355211,0.0003745122,0.0001689823,0.00007512353],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001510222,0.001021574,0.007539346,0.002068783,0.0002856816,0.00001035117,0.01907461,0.0008562421,0.2568922,0.0003649152,0.0008614198,0.7108738],"study_design_scores_gemma":[0.00353148,0.0002408691,0.01003295,0.001095082,0.00006908007,0.00002656613,0.002237516,0.2053256,0.7769106,0.0003013684,0.00001112113,0.0002177445],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07959679,0.00001908605,0.9171916,0.002040617,0.0003174359,0.0003773188,0.00002080259,0.00004838121,0.0003879198],"genre_scores_gemma":[0.9149702,0.00007038218,0.08450593,0.000346263,0.000005983943,0.0000216729,0.000001897484,0.000008632234,0.00006899655],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8353735,"threshold_uncertainty_score":0.3663017,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01084822036847822,"score_gpt":0.2927377472235015,"score_spread":0.2818895268550233,"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."}}