Coronary artery disease (CAD) is the leading cause of death in the United States. The clinical gold standards to diagnose and guide treatment of patients with CAD are based on invasive catheter-based procedures, such as x-ray coronary angiography (XCA) for anatomic assessment or fractional flow reserve (FFR) for physiologic assessment. However, there are costs and risks associated with such invasive procedures. Such concerns are further highlighted by the fact that large studies have shown nearly two-thirds of patients referred for their initial elective invasive XCA were found to have no significant stenoses. Thus, better non-invasive diagnostic tools are needed. Cardiac MRI (CMR) is the only non-invasive imaging modality that provides a comprehensive assessment of CAD in a single examination, including an assessment of myocardial perfusion, cardiac function and viability, as well as angiographic evaluation of stenoses, without requiring ionizing radiation. These properties also allow for repeat testing as may be clinically indicated. However, despite its great potential to serve as the non- invasive gatekeeper for costly invasive procedures, lengthy examination times have prevented CMR from clinical translation. Although several accelerated imaging techniques have been proposed, these still require trade-offs between coverage, resolution and signal-to-noise ratio. In this proposal, we will develop and validate novel acquisition and reconstruction strategies to enable a highly accelerated high-resolution whole heart CMR exam for comprehensive CAD assessment in under 10 minutes. We will develop fast and low specific absorption rate outer volume suppression modules to reduce the source of aliasing artifacts from the chest and the back. This will enable higher rates for simultaneous multi-slice imaging in perfusion and cine CMR, improving coverage substantially with minimal noise amplification. For coronary MRI and viability imaging, simultaneous multi-slab imaging will be introduced to CMR, facilitating high isotropic resolution acquisitions with fast coverage. These acquisitions will be supplemented with regularized leakage-blocking and patient- specific machine learning reconstructions for further artifact and noise removal. Finally, we will implement and validate the proposed rapid comprehensive CMR exam in a cohort of suspected CAD patients, comparing our approach with conventional clinical CMR for the assessment of function, perfusion, and viability, and with invasive XCA for the assessment of coronary stenosis. Successful completion of this project has the potential to transform CMR into a leading rapid non-invasive tool for safe and accurate diagnosis of CAD, improving the healthcare of several million patients with chest pain and other CAD symptoms annually.
Coronary artery disease (CAD) is the leading cause of death in the United States, accounting for one in six deaths. Cardiac MRI is a non-invasive non-ionizing technique for comprehensive evaluation of CAD, but its clinical translation is hampered by lengthy exam times. In this work, we will develop and validate techniques for rapid cardiac MRI in a short exam time for comprehensive CAD assessment.
