{"id":"W4387376141","doi":"10.1038/s41598-023-43715-5","title":"Effects of MRI scanner manufacturers in classification tasks with deep learning models","year":2023,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":48,"is_retracted":false,"has_abstract":true,"ca_institutions":"Women and Children’s Health Research Institute; University of Alberta","funders":"Canadian Institutes of Health Research; Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; Government of Canada; Natural Sciences and Engineering Research Council of Canada; Fondation Brain Canada; ALS Society of Canada","keywords":"Computer science; Deep learning; Artificial intelligence; Scanner; Machine learning; Harmonization; Neuroimaging; Vendor; Medical imaging; Pattern recognition (psychology); Computer vision; Medicine","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":[0.01295768,0.001226216,0.0009059946,0.0008400157,0.0006801403,0.001157057,0.001722174,0.001565189,0.001474777],"category_scores_gemma":[0.03527274,0.0009635891,0.0009280873,0.001269994,0.001266381,0.002397144,0.00249029,0.001396,0.0006838252],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007353082,"about_ca_system_score_gemma":0.0008810553,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00300014,"about_ca_topic_score_gemma":0.00276217,"domain_scores_codex":[0.9924126,0.003372653,0.0009149156,0.001478189,0.001308724,0.0005128745],"domain_scores_gemma":[0.9803657,0.01249112,0.001842546,0.003301261,0.00163252,0.0003668609],"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":[0.01209909,0.001612283,0.1235407,0.0008616747,0.001623596,0.001436472,0.001291879,0.4146039,0.05339681,0.003405283,0.006829744,0.3792985],"study_design_scores_gemma":[0.0006222282,0.00657763,0.08722119,0.0002527141,0.001325529,0.001777138,0.001154481,0.7454001,0.1374556,0.006297487,0.011662,0.0002539507],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.937013,0.001918255,0.05511451,0.0009252899,0.0002883991,0.0001843734,0.0004695092,0.001341306,0.002745342],"genre_scores_gemma":[0.9698889,0.0004160861,0.0262971,0.0005086508,0.00004878628,0.0001274305,0.0009762832,0.0002691436,0.001467508],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01295768,"threshold_uncertainty_score":0.06852758,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01214589929517269,"score_gpt":0.2665665530994395,"score_spread":0.2544206538042668,"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."}}