Rapid Magnetic Resonance Imaging using Advanced Signal Processing Techniques

MRI is a fundamentally slow imaging technique, making the visualization of rapid motion or dynamic processes with MRI difficult. Any motion, including respiration or cardiac motion, can lead to errors in MRI images which can complicate disease diagnosis.  There are still many processes, such as the heart beating in real-time or the flow of blood in the body, that are difficult to visualize even in 2D with MRI.

My laboratory focuses on accelerating MRI scans by collecting data rapidly and reconstructing images using advanced signal processing techniques. We focus on data acquisition along non-Cartesian trajectories, and use both parallel imaging and compressed sensing methods to reconstruct images. By combining these technologies, we are able to perform MRI scans much more rapidly (up to 16x faster) than when using traditional approaches.

While simply speeding up MRI scans is advantageous, we seek to develop rapid imaging methods which can be used to visualize processes that cannot be seen in any other way. We work towards two main applications:

• real-time cardiac imaging, in which the motion of the heart can be assessed without the need for breathholding or EKG gating, which are common in MRI scans
• 3D imaging for abdominal and cardiac scans, where the entire 3D volume can be rapidly assessed

We work closely with clinical colleagues in the Departments of Radiology and Cardiology at University Hospitals of Cleveland to translate the reconstruction methods that we develop into clinical practice.