Objective Assessment of a Step Artifact Correction Algorithm in Sequential Coronary CT Angiography on Dual-Layer CT

Area of Science:

• Medical Imaging
• Cardiovascular Imaging
• Radiology

Background:

• Sequential coronary CT angiography (CCTA) is susceptible to step artifacts due to imaging across different heartbeats.
• These artifacts can obscure coronary plaque and hinder the accuracy of automated analyses like CT-fractional flow reserve.
• Existing methods lack sufficient artifact reduction, impacting diagnostic capabilities.

Purpose of the Study:

• To evaluate a vendor-specific step artifact reduction (SAR) algorithm for sequential CCTA.
• To assess the combined efficacy of SAR and motion-compensated reconstruction (MCR) in improving image quality.
• To quantify the impact of these algorithms on lumen continuity and artifact reduction.

Main Methods:

• Retrospective analysis of 79 patients undergoing sequential CCTA on a 128-slice dual-layer spectral CT system.
• Comparison of three reconstruction methods: clinical standard, SAR alone, and SAR combined with MCR.
• Quantification of lumen continuity using Jaccard index (JI) across step interfaces and histogram entropy of subtraction images for whole-heart continuity.

Main Results:

• SAR significantly increased mean Jaccard index (JI) across 165 coronary step segments from 0.38 to 0.58 (P < 0.01), indicating improved lumen continuity.
• Histogram entropy decreased with SAR (3.81 to 3.51; P < 0.01), signifying reduced step-related artifacts.
• In motion-affected segments, SAR+MCR achieved a higher JI (0.70) compared to SAR alone (0.49; P < 0.01).

Conclusions:

• The vendor-specific SAR algorithm effectively improves vessel and whole-heart continuity in sequential CCTA on dual-layer spectral CT.
• Combining SAR with MCR provides additional benefits for image quality in motion-affected segments.
• These algorithmic enhancements hold promise for reducing artifacts and improving the reliability of automated CCTA analyses.