Improvement of Image Quality and Diagnostic Performance by an Innovative Motion-Correction Algorithm for Prospectively ECG Triggered Coronary CT Angiography

• Published in: PloS one Vol. 10; no. 11; p. e0142796
• Main Authors: Li, Zhen-Nan, Yin, Wei-Hua, Lu, Bin, Yan, Hong-Bing, Mu, Chao-Wei, Gao, Yang, Hou, Zhi-Hui, Wang, Zhi-Qiang, Liu, Kun, Parinella, Ashley H., Leipsic, Jonathon A.
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 16.11.2015; Public Library of Science (PLoS)
• Subjects: Adult; Aged; Algorithms; Angiography; Cardiovascular disease; CAT scans; Comparative analysis; Computed tomography; Coronary Angiography; Coronary artery; Diagnostic systems; Echocardiography; EKG; Electrocardiography; Female; Health aspects; Heart rate; Humans; Image quality; Male; Medical diagnosis; Medical imaging; Middle Aged; Motion; Patients; Radiographic Image Interpretation, Computer-Assisted; Reconstruction; ROC Curve; Statistical analysis; Statistical significance; Stenosis; Tomography; Tomography, X-Ray Computed; United States; Beijing China; China
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0142796

To investigate the effect of a novel motion-correction algorithm (Snap-short Freeze, SSF) on image quality and diagnostic accuracy in patients undergoing prospectively ECG-triggered CCTA without administering rate-lowering medications. Forty-six consecutive patients suspected of CAD prospectively underwent CCTA using prospective ECG-triggering without rate control and invasive coronary angiography (ICA). Image quality, interpretability, and diagnostic performance of SSF were compared with conventional multisegment reconstruction without SSF, using ICA as the reference standard. All subjects (35 men, 57.6 ± 8.9 years) successfully underwent ICA and CCTA. Mean heart rate was 68.8±8.4 (range: 50-88 beats/min) beats/min without rate controlling medications during CT scanning. Overall median image quality score (graded 1-4) was significantly increased from 3.0 to 4.0 by the new algorithm in comparison to conventional reconstruction. Overall interpretability was significantly improved, with a significant reduction in the number of non-diagnostic segments (690 of 694, 99.4% vs 659 of 694, 94.9%; P<0.001). However, only the right coronary artery (RCA) showed a statistically significant difference (45 of 46, 97.8% vs 35 of 46, 76.1%; P = 0.004) on a per-vessel basis in this regard. Diagnostic accuracy for detecting ≥50% stenosis was improved using the motion-correction algorithm on per-vessel [96.2% (177/184) vs 87.0% (160/184); P = 0.002] and per-segment [96.1% (667/694) vs 86.6% (601/694); P <0.001] levels, but there was not a statistically significant improvement on a per-patient level [97.8 (45/46) vs 89.1 (41/46); P = 0.203]. By artery analysis, diagnostic accuracy was improved only for the RCA [97.8% (45/46) vs 78.3% (36/46); P = 0.007]. The intracycle motion correction algorithm significantly improved image quality and diagnostic interpretability in patients undergoing CCTA with prospective ECG triggering and no rate control.