Coronary magnetic resonance angiography (MRA) is emerging as a promising method for noninvasive diagnosis of coronary artery diseases. Nonetheless, acquisition of large-coverage high-resolution coronary artery images requires free-breathing acquisition. Accurately measuring and correcting respiratory motion has been one of the major challenges to coronary MRA. Diaphragmatic navigator (NAV), the current state-of-art technique, indirectly estimates heart motion along the heart-feet direction only and therefore suffers from low acquisition efficiency and residual motion artifacts. Slow-contrast infusion was recently proposed to enable contrast-enhanced whole-heart coronary MRA. However, synchronization of data acquisition with contrast enhancement becomes more complicated than conventional protocols and remains unresolved. The improper acquisition scheme currently in use results in low signal-to-noise ratio (SNR) and therefore suboptimal visualization of coronary arteries. This work was aimed to improve the robustness of coronary MRA by better resolving the aforementioned difficulties. To address the inaccuracy of NAV, a novel respiratory self-gating (RSG) approach was developed. Respiratory motion was directly monitored in a heart projection acquired in the imaging volume. In our quantitative and qualitative evaluations, RSG demonstrated its superiority to NAV in motion detection accuracy and its potential for improving coronary artery delineation. Furthermore, this work also demonstrated the feasibility of using this RSG method to resolve respiratory motion for cardiac-motion-resolved 3D coronary MRA. To address the residual motion with 1D motion detection, RSG was further extended to encompass 3D translation detection (3D RSG). By repeatedly acquiring a few RSG echoes, breathing-induced heart displacements during a scan could be directly measured and compensated. With more complete motion correction, 3D RSG further improved image quality as well as acquisition efficiency for free-breathing coronary MRA. Additionally, to address the current difficulty in contrast-enhanced whole-heart coronary MRA, a auto-tracking technique was proposed, capable of automatically triggering data acquisition at the proper time. With auto-tracking, both SNR and visual vessel delineation were significantly improved. Further work developed contrast-kinetics-resolved coronary MRA by combining temporal sliding window with time-resolved 3D projection reconstruction. A series of time frames could be reconstructed, capturing the contrast kinetics during a scan. Those frames with optimal coronary artery visualization could be selected automatically for diagnosis.

Last modified

• 09/20/2018

Creator

• Peng Lai

DOI

• https://doi.org/10.21985/N2MF1Z

Subject

• Biomedical Engineering

Keyword

• Coronary angiography
• Motion correction
• Magnetic resonance imaging
• Contrast agent
• Eespiratory gating

Date created

• 2008-11-11

Resource type

• Dissertation

Rights statement

• In Copyright