Congratulations to Roshan D’Souza, associate professor, mechanical engineering, who was awarded a three-year, $298,500 National Science Foundation grant to support his research to advance the newest generation of MRI technology.
Magnetic Resonance Imaging (MRI) is rapidly evolving; 4D MRI adds a time dimension to this traditionally 3D technology, making it possible to image and analyze biological processes, such as blood flow (or hemodynamics) over time and track variations.
However, there are hurdles on the road to utilizing the newest technology. Accurate and reliable hemodynamic analysis of cardio-vascular flows from 4D-flow MRI data are time consuming and difficult, sometimes impossible, due to factors including low spatio-temporal resolution, noise, and image artifacts, says D’Souza.
Working with researchers at Northern Arizona University, D’Souza aims to use physics informed deep learning (a type of machine learning) to address these three issues and create images that are sharp, clear and free of electronic noise.
The project—entitled “Collaborative Research: Enhanced 4D-Flow MRI through Deep Data Assimilation for Hemodynamic Analysis of Cardiovascular Flows” – will provide opportunities to train graduate students at UWM and NAU in the latest deep-learning based techniques used in engineering, engage undergraduate students in research at both universities, and engage Milwaukee high school students, many of whom belong to marginalized communities, through UWM’s summer programs.
Forty-eight percent of adult Americans have cardiovascular diseases, which include atherosclerosis, aneurysms, and congestive heart failure. Hemodynamic forces play a critical role in the development of these conditions. For example, the forces related to blood’s interaction with the walls of an artery have a major impact on the initiation and progression of aneurysms, atherosclerosis, and vasospasms.
The research builds on D’Souza’s previous work, in collaboration with the Medical College of Wisconsin, on new methods of analysis for 4D-Flow MRI to support diagnosis and treatment of brain aneurysms.