Machine Learning for Brain Tumors
Dr. Xue Feng, Research Assistant Professor of BME and CTO of local BME start-up Springbok Analytics, was one of two first place finishers in the BraTS 2018 Challenge. The goal of the competition was to identify the best machine learning algorithms for brain tumor segmentation, progression assessment, and overall survival prediction. Feng finished first in the survival prediction task, winning a cash prize and an invitation to speak on the topic this fall in Grenada, Spain.
The BraTS challenge focuses on the evaluation of state-of-the-art methods for the segmentation of brain tumors in multimodal magnetic resonance imaging (MRI) scans.
About Xue Feng, PhD
As a student in Craig Meyer's lab, Feng did extensive research using the spiral k-space trajectory on both sequence and reconstruction development in cardiac and speech MRI. After earning his PhD in 2012, Feng has continued to work on dynamic MR imaging, and he's also added projects on MR thermometry and muscle MRI. To meet the need for extensive image processing such as segmentation, Feng gained expertise in automatic muscle segmentation for accurate quantification.
More recently, Feng has gained extensive experience in deep learning due to its superiority in many image processing tasks. He's participated in several challenge competition with good results. Feng currently focuses on using deep learning in multiple image segmentation tasks, including cardiac MRI for hypertrophic cardiomyopathy, individual muscle segmentation and organ segmentation based on CT. Feng is CTO of Springbok Analytics and Research Assistant Professor of Biomedical Engineering. Springbok Analytics spun out of the University of Virginia, based on multiple years' research on advanced MRI, biomechanics and sports injury.
Springbok Analytics received early funding from the UVA-Coulter Translational Research Partnership.