{"id":"W4387267741","doi":"10.36227/techrxiv.24198324","title":"Clean self-supervised MRI reconstruction from noisy, sub-sampled training data with Robust SSDU","year":2023,"lang":"en","type":"preprint","venue":"","topic":"Advanced MRI Techniques and Applications","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Sunnybrook Hospital; University of Toronto","funders":"Engineering and Physical Sciences Research Council; Royal Academy of Engineering; National Institute for Health and Care Research; Canada Research Chairs; Wellcome Trust","keywords":"Undersampling; Computer science; Artificial intelligence; Noise (video); Benchmark (surveying); Pattern recognition (psychology); Noise reduction; Deep learning; Inference; Machine learning; Image (mathematics)","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001650824,0.001781922,0.001382611,0.0005735264,0.0003790798,0.00105555,0.002190821,0.001762404,0.003254723],"category_scores_gemma":[0.004656617,0.0007567753,0.0009888574,0.0006647719,0.001089907,0.001425741,0.001639665,0.002012402,0.002200769],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000687027,"about_ca_system_score_gemma":0.001316297,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005369975,"about_ca_topic_score_gemma":0.009726285,"domain_scores_codex":[0.9991352,0.0002258915,0.00003993623,0.0002518457,0.0002778627,0.0000693722],"domain_scores_gemma":[0.9983945,0.0004573542,0.0001487537,0.0005777935,0.0003421536,0.0000793411],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004833965,0.0002654322,0.001428592,0.0002800808,0.0002576854,0.0003301468,0.000149677,0.4056674,0.04211799,0.007034525,0.02072073,0.5212644],"study_design_scores_gemma":[0.00001228247,0.00004711046,0.000200616,0.000009092907,0.000007151064,0.0000689647,0.000009283547,0.9874774,0.008705759,0.002133772,0.001316071,0.00001239348],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02080396,0.0006701641,0.9697003,0.0004708758,0.000110426,0.00009092816,0.0004000654,0.005339369,0.002413864],"genre_scores_gemma":[0.2404732,0.0006570807,0.7418126,0.0008287833,0.0001389993,0.000230837,0.003666908,0.0009719889,0.01121963],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005369975,"threshold_uncertainty_score":0.0108881,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1526329694512915,"score_gpt":0.3333794641213845,"score_spread":0.180746494670093,"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."}}