{"id":"W4413049956","doi":"10.1016/j.jcmg.2025.06.010","title":"General Purpose Deep Learning Attenuation Correction Improves Diagnostic Accuracy of SPECT MPI","year":2025,"lang":"en","type":"article","venue":"JACC. Cardiovascular imaging","topic":"Cardiac Imaging and Diagnostics","field":"Medicine","cited_by":5,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Ottawa; University of Calgary","funders":"National Heart, Lung, and Blood Institute","keywords":"Coronary artery disease; Medicine; Correction for attenuation; Myocardial perfusion imaging; Single-photon emission computed tomography; Nuclear medicine; Cohort; Receiver operating characteristic; Area under the curve; Emission computed tomography; Radiology; Internal medicine; Positron emission tomography","routes":{"ca_aff":true,"ca_fund":false,"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":[0.0007751309,0.001452702,0.0007556565,0.0006805592,0.0002958973,0.0008855247,0.001094335,0.00104075,0.007914432],"category_scores_gemma":[0.002996512,0.0004063268,0.0007048542,0.0004540464,0.0002297596,0.0006259487,0.001149789,0.001223665,0.003055298],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003891312,"about_ca_system_score_gemma":0.001008891,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006041165,"about_ca_topic_score_gemma":0.009676424,"domain_scores_codex":[0.9997628,0.00004807965,0.00001303961,0.00007184194,0.00005583089,0.0000484432],"domain_scores_gemma":[0.9994745,0.0001748202,0.00004704283,0.00007879041,0.0001831626,0.00004170885],"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":[0.0008942811,0.0003083633,0.006955804,0.0003265397,0.0002263676,0.0002128844,0.00003588049,0.07291059,0.03464229,0.001040824,0.02902348,0.8534228],"study_design_scores_gemma":[0.00009998794,0.0002189986,0.005973141,0.00006631705,0.0001634365,0.0004186465,0.00002341838,0.9561741,0.02699648,0.003343959,0.006478419,0.00004315209],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2145739,0.006854361,0.7307879,0.003927321,0.001053268,0.0002104786,0.005043727,0.02475563,0.01279335],"genre_scores_gemma":[0.796515,0.001912657,0.171747,0.001269614,0.0005977569,0.0001454591,0.006551296,0.001264641,0.01999657],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007914432,"threshold_uncertainty_score":0.02647638,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006170189341469448,"score_gpt":0.2546587565012315,"score_spread":0.248488567159762,"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."}}