{"id":"W3014584416","doi":"10.7488/era/239","title":"Development of machine learning schemes for segmentation, characterisation, and evolution prediction of white matter hyperintensities in structural brain MRI","year":2020,"lang":"en","type":"dissertation","venue":"ERA","topic":"Medical Image Segmentation Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian Institutes of Health Research; National Institutes of Health; UK Dementia Research Institute; Lembaga Pengelola Dana Pendidikan; Eisai; Fondation Leducq; Mrs Gladys Row Fogo Charitable Trust; BioClinica; European Commission; U.S. Department of Defense; Alzheimer's Disease Neuroimaging Initiative; University of Southern California; Wellcome Trust; University of Edinburgh; Northern California Institute for Research and Education; Nvidia; Bristol-Myers Squibb; Eli Lilly and Company; Biogen; Alzheimer's Association","keywords":"Hyperintensity; White matter; Artificial intelligence; Segmentation; Pattern recognition (psychology); Computer science; Magnetic resonance imaging; Medicine; Radiology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002119096,0.0001470794,0.000254485,0.0002195003,0.0000821799,0.00002793427,0.0001454256,0.00009268763,0.00003258083],"category_scores_gemma":[0.00009137825,0.0001514452,0.00003233262,0.000158659,0.0000291465,0.0003689312,0.00005087497,0.0001607804,8.346173e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007251582,"about_ca_system_score_gemma":0.0001154157,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001266418,"about_ca_topic_score_gemma":0.00002017314,"domain_scores_codex":[0.9986983,0.00005384754,0.0006114694,0.0002725649,0.0002580444,0.00010576],"domain_scores_gemma":[0.9991475,0.00004847251,0.0004404694,0.00009873783,0.0002281114,0.00003673371],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.0003264667,0.00008559845,0.1169317,0.004605827,0.0001935505,0.000002608143,0.181916,0.00003063392,0.5857108,0.001845141,0.001185386,0.1071663],"study_design_scores_gemma":[0.0009373454,0.0001577657,0.657239,0.0004772208,0.00002160642,0.000004690821,0.00262106,0.03381274,0.3035385,0.0004180196,0.0004709341,0.0003011541],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.3910109,0.00008483837,0.6076549,0.0003836548,0.0002131281,0.0005225906,0.00001971003,0.00006452097,0.00004573277],"genre_scores_gemma":[0.3970661,0.00001985301,0.5997292,0.0001729982,0.00004125302,0.0001376442,0.002153823,0.00002246723,0.0006567058],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.5403073,"threshold_uncertainty_score":0.6175758,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01053361147876579,"score_gpt":0.2575088281232527,"score_spread":0.2469752166444869,"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."}}