A professor from Old Dominion University has received more than $2 million from the National Institutes of Health to improve how doctors detect and treat glioblastoma — the most aggressive and deadly form of brain tumor.
Khan Iftekharuddin, who teaches in ODU’s department of electrical and computer engineering, is leading the multi-year project. His team is developing artificial intelligence models to detect glioblastoma recurrence and identify highly aggressive tumor subtypes without the need for invasive biopsies.
“Out of hundreds of MRI images collected for each patient, only a few slices actually show signs of the tumor,” Iftekharuddin said. “We looked at how we can process this data, develop computational models to automatically analyze these images, remove the ones that do not need to be looked over, and then suggest the ones that clinicians should review.”
Glioblastoma typically grows rapidly and often returns even after surgery or chemotherapy. Iftekharuddin said patients usually survive less than two years after diagnosis.
Data from the American Brain Tumor Association shows glioblastoma accounts for roughly 14 % of all primary brain tumors in the U.S., and more than 12,000 cases are diagnosed each year nationwide.
In Virginia, data for brain and central nervous system cancers, which include glioblastoma, reveal an age-adjusted incidence of about 6.2 new cases per 100,000 people from 2010-2014.
Collaborating with hospitals nationwide by combining MRI scans and clinical data, Iftekharuddin’s team hopes to give doctors a faster, more precise way to identify when a tumor is coming back and how aggressive it might be.
That could help doctors tailor surgery or chemotherapy plans more precisely — leading to more personalized care and better quality of life for patients facing aggressive brain cancers.
“The current focus is really assisting physicians to come up with better patient management,” Iftekharuddin said. “If clinicians know from this analysis that a patient may have a more aggressive form of the tumor, then they can adjust their treatment plan accordingly.”
The $2 million funding will be used to expand data collection across multiple hospitals, support graduate students and postdoctoral researchers and further develop AI algorithms capable of analyzing complex medical images and clinical information.
Iftekharuddin’s team also wants to build a robust database that can be shared among research institutions for future studies.
The grant builds on Iftekharuddin’s 2016 award from the National Institute of Biomedical Imaging and Bioengineering. That research focused on developing more accurate methods to model, analyze and segment brain tumors in order to track their growth, classify tumor types and predict patient survival rates.
