Opportunistic deep learning powered calcium scoring in oncologic patients with very high coronary artery calcium (≥ 1000) undergoing 18F-FDG PET/CT

• DOI: 10.1038/s41598-022-20005-0
• Publication Date: 2022-11-10T11:03:26Z
• Authors: Elisabeth Sartoretti; Antonio G. Gennari; Alexander Maurer; Thomas Sartoretti; Stephan Skawran; Moritz Schwyzer; Alexia Rossi; Andreas A. Gianno...; Ronny R. Buechel; Catherine Gebhard; Martin W. Huellner; Michael Messerli
• Keywords: Agatston score

Abstract Our aim was to identify and quantify high coronary artery calcium (CAC) with deep learning (DL)-powered CAC scoring (CACS) in oncological patients known very (≥ 1000) undergoing 18F-FDG-PET/CT for re-/staging. 100 were enrolled: 50 Agatston scores ≥ 1000 (high CACS group), < (negative control group). All underwent cardiac SPECT myocardial perfusion imaging (MPI) by 99mTc-tetrofosmin within 6 months. manually performed on dedicated non-contrast ECG-gated CT scans obtained from SPECT-MPI (reference standard). Additionally, fully automatically a user-independent DL-CACS tool non-contrast, free-breathing, non-gated examinations. Image quality noise of assessed. manual DL compared. The group had 2200 ± 1620 standard) 1300 1011 (DL tool, average underestimation 38.6 26%) an intraclass correlation 0.714 (95% CI 0.546, 0.827). Sufficient image significantly improved the tool’s capability correctly assigning ( p = 0.01). In group, assigned all cases. conclusion, DL-based examinations underestimates load, but assigns over 70% cases, provided sufficient quality. Subgroup analyses showed that does not generate false-positives.