Detection of Coronary Artery Inflammation Using Dual Energy CT
Presentations
PO-GePV-I-97 (Sunday, 7/10/2022) [Eastern Time (GMT-4)]
ePoster Forums
Purpose: Early detection of vascular inflammation - a key feature in atherogenesis - would allow better cardiovascular risk stratification to prevent or treat different disease states. Perivascular adipose tissue is primarily composed of water and lipid. Vascular inflammation is expected to increase the water component of perivascular adipose tissue. This simulation study is to investigate the feasibility of distinguishing diseased and healthy coronary arteries by quantifying the chemical composition of perivascular adipose tissue in terms of water and lipid using dual-energy CT.
Methods: A CT simulation package was developed to match the physical parameters of a clinical 320-slice CT scanner. Digital phantoms with mixture of known volumetric fraction of water and lipid from 0% to 100% were used for calibration. A total of 10 digital calibration phantoms were scanned at 80, 120 and 135 kVp using the simulation package. The expected clinical range of water volume fraction was determined at 120 kVp based on the previous clinical studies of human perivascular fat. Dual-energy material decomposition technique was implemented to validate the volumetric fraction of water within the clinical range.
Results: The expected clinical range of water volume fraction was calculated to be 30%-50%. The measured and known water percentages were correlated by PM = 1.00PK – 1.22E-6 (r=1.00) with negligible root mean-squared error (RMSE).
Conclusion: The simulation results show that dual-energy material decomposition can accurately quantify the water-lipid composition of perivascular fat. With accurate quantification, this decomposition method can potentially detect coronary artery inflammation early and enable deployment of targeted treatment.
Funding Support, Disclosures, and Conflict of Interest: UCI Undergraduate Research Opportunities Program (UROP)
Keywords
Simulation, Dual-energy Imaging, Quantitative Imaging
Taxonomy
IM- CT: Dual Energy and Spectral
Contact Email