Abstract

To assess the effect of a second-generation motion correction algorithm on the diagnostic accuracy of coronary computed tomography angiography (CCTA) using a 256-detector row CT in patients with increased heart rates. Eighty-one consecutive symptomatic cardiac patients with increased heart rates (≥ 75 beats per min) were enrolled. All patients underwent CCTA and invasive coronary angiography (ICA). CCTA was performed with a 256-detector row CT using prospectively ECG-triggered single-beat protocol. Images were reconstructed using standard (STD) algorithm, first-generation intra-cycle motion correction (MC1) algorithm, and second-generation intra-cycle motion correction (MC2) algorithm. The image quality of coronary artery segments was assessed by two experienced radiologists using a 4-point scale (1: non-diagnostic and 4: excellent), according to the 18-segment model. Diagnostic performance for segments with significant lumen stenosis (≥ 50%) was compared between STD, MC1, and MC2 by using ICA as the reference standard. The mean effective dose of CCTA was 1.0mSv. On per-segment level, the overall image quality score and interpretability were improved to 3.56 ± 0.63 and 99.2% due to the use of MC2, as compared to 2.81 ± 0.85 and 92.5% with STD and 3.21 ± 0.79 and 97.2% with MC1. On per-segment level, compared to STD and MC1, MC2 improved the sensitivity (92.2% vs. 79.2%, 80.7%), specificity (97.8% vs. 82.1%, 90.8%), positive predictive value (89.9% vs. 48.4%, 65.1%), negative predictive value (98.3% vs. 94.9%, 95.7%), and diagnostic accuracy (96.8% vs. 81.5%, 89.0%). A second-generation intra-cycle motion correction algorithm for single-beat CCTA significantly improves image quality and diagnostic accuracy in patients with increased heart rate. • A second-generation motion correction (MC2) algorithm can further improve the image quality of all coronary arteries than a first-generation motion correction (MC1). • MC2 algorithmcan significantly reduce the number of false positive segments compared to standard and MC1 algorithm.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call