Clinical evaluation of new automatic coronary-specific best cardiac phase selection algorithm for single-beat coronary CT angiography

Abstract

The aim of this study was to evaluate the workflow efficiency of a new automatic coronary-specific reconstruction technique (Smart Phase, GE Healthcare-SP) for selection of the best cardiac phase with least coronary motion when compared with expert manual selection (MS) of best phase in patients with high heart rate. A total of 46 patients with heart rates above 75 bpm who underwent single beat coronary computed tomography angiography (CCTA) were enrolled in this study. CCTA of all subjects were performed on a 256-detector row CT scanner (Revolution CT, GE Healthcare, Waukesha, Wisconsin, US). With the SP technique, the acquired phase range was automatically searched in 2% phase intervals during the reconstruction process to determine the optimal phase for coronary assessment, while for routine expert MS, reconstructions were performed at 5% intervals and a best phase was manually determined. The reconstruction and review times were recorded to measure the workflow efficiency for each method. Two reviewers subjectively assessed image quality for each coronary artery in the MS and SP reconstruction volumes using a 4-point grading scale. The average HR of the enrolled patients was 91.1±19.0bpm. A total of 204 vessels were assessed. The subjective image quality using SP was comparable to that of the MS, 1.45±0.85 vs 1.43±0.81 respectively (p = 0.88). The average time was 246 seconds for the manual best phase selection, and 98 seconds for the SP selection, resulting in average time saving of 148 seconds (60%) with use of the SP algorithm. The coronary specific automatic cardiac best phase selection technique (Smart Phase) improves clinical workflow in high heart rate patients and provides image quality comparable with manual cardiac best phase selection. Reconstruction of single-beat CCTA exams with SP can benefit the users with less experienced in CCTA image interpretation.

Fig 1. Comparison of subjective image quality of MS and SP phases.

MS = Manual Selection, SP = Smart Phase. Image quality score: 1 = excellent, 2 = good, 3 = intermediate, 4 = poor.

Fig 2. Comparison of the numbers of reconstructions and time by MS and SP algorithms.

MS = Manual Selection, SP = Smart Phase.

Fig 3. Plots of selected cardiac phases.

Distribution of selected cardiac phases with the use of MS and SP methods. MS-manual selection, SP-smart phase.

Fig 4. The box plot of phase reconstructions.

The center phase of the two groups were almost overlap, but the SP group were more scatted than the MS group.

Fig 5. Differences in the selected phases.

There was no statistic difference between the two groups.