Research

Deep Learning for Quantitative Assessment of Hypertrophic Cardiomyopathy

Hypertrophic cardiomyopathy (HCM) is the most common monogenic heart disease.  HCM is characterized by unexplained left ventricular hypertrophy (LVH), myofibrillar disarray, and myocardial fibrosis. There is a pressing need for rapid and robust quantification of cardiac MR markers for identification of risk.  Currently the quantification relies heavily on manual image segmentation of LV, RV and scar regions, which is not only time-consuming but also suffers from significant inter-observer variability in multi-site studies. Deep learning methods have recently shown promising results in image-related tasks including medical image segmentation, because they can handle complicated imaging situations such as variations in intensity, contrast and shape. The goal of this study is to develop deep-learning automatic image analysis methods to greatly improve the efficiency and robustness of quantitative CMR markers of disease.

MRI Guidance of Focused Ultrasound Neurosurgery

MR-guided focused ultrasound surgery (FUS) shows substantial promise for minimally invasive neurosurgery.  UVa neurosurgeons are world leaders in FUS treatment of essential tremor.  While the current methods for MRI guidance of focused ultrasound surgery are important and valuable, many improvements in MRI guidance are possible and would make the technique a more robust therapeutic modality.   Our current research is focused on developing better methods of temperature monitoring, methods to directly measure tissue ablation, and methods for monitoring unintended heating of the brain.  Our long-term goal is to provide more comprehensive MRI feedback for intracranial FUS procedures. This will include methods to improve the efficiency and safety of intracranial FUS, ensure the treatment is effective, and expand the number of patients that can be effectively treated.

CrCEST in the study of creatine metabolism kinetics in Peripheral Arterial Disease

Description to be updated

High Resolution Real-time Spiral Cine with Whole Heart Coverage

Description to be updated

Rapid Pediatric MRI without Sedation

The need for sedation is a significant concern for magnetic resonance imaging (MRI) of young children. This challenge arises because the typically long durations of volumetric imaging used in clinical applications are too long for infants and young children to remain still voluntarily. Without anesthesia, the motion during a scan would distort images, obscuring critical features needed for diagnosis. Breathing-related motion further reduces image quality, and breath-holds may be unrealistic for young children.  This project aims to eliminate the use of sedation from many pediatric MRI scans using novel acquisition and reconstruction methods that accelerate the imaging process and track and compensate for motion. A pediatric MRI paradigm performed without sedation would enable broader application of clinical MRI in young children. These benefits should make MRI a more attractive option for pediatric imaging versus an alternative such as x-ray computed tomography (CT), which involves a significant radiation dose. This distinction becomes even more important when repeated imaging is necessary, such as when monitoring solid tumors in pediatric cancer patients.