Motion-Compensated Contrast-Enhanced Cardiac CT for Perfusion and Angiography

Taguchi, Katsuyuki

Abstract

Cardiovascular diseases remain the leading cause of death in the western world, placing an ever-increasing burden on both private and public health services. Electrocardiogram (ECG)-gated coronary computed tomography angiography (CTA) imaging is an established non-invasive technique for detecting coronary stenosis caused by calcium deposits and fatty soft atherosclerosis. Thanks to few false negatives, CTA's negative predictive value (NPV) is high enough (>90%) to decrease the number of diagnostic catheter coronary angiography procedures that show no stenosis. A recent study, however, found that twice as many patients were sent for invasive cardiac procedures after CTA assessment as after stress myocardial perfusion scintigraphy. The reason for this is the low positive predictive value (PPV) of CTA due to a relatively large fraction of false positives. It is therefore desirableto combine CTA and a stress test to decrease false positives. Although other modalities can be used for a stress test, the use of stress myocardial CT perfusion (CTP) is desirable because CTP is low cost and has an efficient workflow, and it eliminates problems such as misregistration associated with multimodality imaging. There are the following challenges with CTP and this proposal addresses them in order to establish CTP as a clinical routine: (1) the large uncertainty of pixel values due to so called halfscan artifacts; which may result in (2) lo reproducibility of exam results; (3) the insufficient signal difference (or contrast) between ischemic lesions and healthy tissues; and (4) radiation dose to patients. The problems described above can be attributed to the fact that the current cardiac image reconstruction method used by commercial CT scanners, the halfscan algorithm, does not compensate for the cardiac motion. We propose to solve these problems with CTP by compensating for cardiac motion. Specifically, we propose the following scan and image reconstruction methods to address the problems and we will optimize parameters for detection/characterization tasks for CTP and CTA: (1) Motion estimation and motion compensated image reconstruction (ME-MCR); (2) ECG-gated tube current modulation; (3) single-energy with low kV or dual-energy scan; and (4) larger reconstruction window, wR. The central hypothesis of this project is that ME-MCR with low kV or dual-energy will substantially improve detection and characterization of perfusion defects with CTP and of stenosis with CTA. We expect that the proposed method will improve the positive predictive value from the current 58% to 95% (at a negative predictive value of 95%). Our preliminary studies showed that early versions of the proposed method improve the performance of CTP. The goals are within striking distance and can be achieved by enhancing the tissue contrast and decreasing noise using low kV, dual-energy, and iterative MCR.

Public Health Relevance

This project aims to improve imaging tests used to determine whether coronary arteries are blocked by increasing the sensitivity of stress myocardial x-ray computed tomography perfusion imaging tests. Combining the improved stress test with coronary computed tomography angiography, which has good specificity, will produce a test that is very accurate, cost effective, and fast and will reduce the needed for invasive heart procedures.