Bilyana Tzolova, a third-year doctoral student in computational and applied mathematics (CAAM), has been awarded a three-year, $138,000 predoctoral fellowship by the National Institutes of Health to develop a more accurate approach to imaging the vascular system.
Her grant is titled “Machine Learning Inspired Physical Models in Organs.”
“The vascular system plays a crucial role in diagnosing and treating many diseases,” Tzolova said. “Currently, practitioners locate vessels manually on each image of a CT scan. This is tedious and the process varies highly depending on the individual’s experience and ability.”
Tzolova is researching a way to automate this process using a method known as vessel segmentation, focusing on the liver as the vessels and hepatic tissue are difficult to differentiate from one another. A novel neural network algorithm shows potential for reducing training time and increasing accuracy.
“We will create a model that simulates blood flow and solute transport in the vascular system of the liver. We’ll do this by using coupled multidimensional computational models for flow and transport within the blood vessels. That combination will give a complete overview of the location and function of a patient’s circulatory system,” she said.
Tzolova is advised by Béatrice Rivière, Noah Harding Chair and Professor of CAAM at Rice, and David Fuentes, associate professor of imaging physics at the University of Texas M.D. Anderson Cancer Center in Houston. Tzolova earned her B.S. in physics and her M.S. in mathematics, both from Johns Hopkins University in 2016.