Automatic Quantification of Local Plaque Thickness Differences as Assessed by Serial Coronary Computed Tomography...

Area of Science:

• Cardiovascular Imaging
• Radiology
• Medical Diagnostics

Background:

• Serial coronary computed tomography angiography (CCTA) enables early assessment of coronary plaque progression, vital for preventing major adverse cardiac events (MACEs).
• Traditional CCTA analysis relies on anatomical landmarks for scan matching, which can be imprecise.
• Automated tools for quantifying local plaque thickness differences in serial CCTA have recently emerged.

Purpose of the Study:

• To establish thresholds for plaque thickness differences indicating progression or regression in serial CCTA.
• To determine how these thresholds are influenced by the contrast-to-noise ratio (CNR).

Main Methods:

• A cohort of 50 patients with chest complaints underwent serial CCTA.
• Plaque contours were semi-automatically delineated in 300 coronary vessels.
• Regions of interest were manually drawn to quantify CNR for each vessel.
• Linear regression analysis was performed on plaque thickness differences and CNR to develop calibration graphs.

Main Results:

• A linear regression model was developed correlating CNR with plaque thickness differences.
• Thresholds were identified: plaque progression above 1.009 - 0.002 × CNR.
• Thresholds were identified: plaque regression below -1.638 + 0.012 × CNR.

Conclusions:

• Feasible development of vessel-specific, quality-based thresholds for local plaque thickness differences in serial CCTA.
• These thresholds can potentially enhance early detection of atherosclerosis progression.
• This approach may improve the management of cardiovascular disease risk.