{"id":"W3156072047","doi":"10.59275/j.melba.2021-df47","title":"Adversarial Robust Training of Deep Learning MRI Reconstruction Models","year":2021,"lang":"en","type":"article","venue":"The Journal of Machine Learning for Biomedical Imaging","topic":"Advanced X-ray and CT Imaging","field":"Engineering","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institutes of Health; National Institute of Arthritis and Musculoskeletal and Skin Diseases; University of California, San Francisco; York University","keywords":"Computer science; Artificial intelligence; Iterative reconstruction; Deep learning; Generalization; Machine learning; Workflow; Set (abstract data type); Fidelity; 3D reconstruction; Code (set theory); Pattern recognition (psychology); Computer vision; Mathematics","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":[0.001379843,0.000188948,0.0004022083,0.0001899071,0.0002428349,0.00003197882,0.0002355167,0.00006133567,0.00006632713],"category_scores_gemma":[0.0005941556,0.0001511712,0.0002103031,0.0003159251,0.0001837958,0.0004476696,0.00005278515,0.001223943,0.000001166388],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006861451,"about_ca_system_score_gemma":0.00006091647,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000006240089,"about_ca_topic_score_gemma":0.000001689889,"domain_scores_codex":[0.9982752,0.0001809247,0.0006993494,0.0001226541,0.0003623412,0.0003595159],"domain_scores_gemma":[0.998664,0.0005104619,0.0003616732,0.0001149904,0.0002200257,0.0001289116],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00006729867,0.00001446248,0.0002170148,0.00006743881,0.00009600516,0.00001605701,0.001674224,0.733999,0.01088131,0.00006954763,0.00002447288,0.2528732],"study_design_scores_gemma":[0.001272153,0.00007179569,0.00004356822,0.0002400534,0.0001420279,0.001232972,0.003396674,0.9861519,0.001691428,0.001518966,0.004070223,0.0001682477],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02744781,0.003527994,0.9665939,0.0006207668,0.0009553058,0.00006944958,0.000002808999,0.0000854007,0.0006966197],"genre_scores_gemma":[0.9625588,0.0004967222,0.03622645,0.00003936394,0.0005563212,0.000001678244,0.00001347481,0.00005515417,0.0000520381],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.935111,"threshold_uncertainty_score":0.6164584,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0130247580965794,"score_gpt":0.2346587911582449,"score_spread":0.2216340330616655,"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."}}