The National Institute of Health awarded a $1.7 million grant to a research team at the University for further development of new computational tools. Grant recipients include Mark Anastasio, head of the Department of Bioengineering, and Frank Brooks, Hua Li and David Forsyth, professors in Engineering. Mohammad Eghtedari, professor at the University of California, San Diego, is also on the team.
For the last five years, the team of computational researchers has utilized deep learning methods to extract information from images. Namely, their work analyzes the quality and usefulness of images for different purposes.
For an image to be useful in gathering information about a patient, it has to be visually clear. Bioimaging technologies, such as ultrasounds, CT scans and X-rays, are used by technicians with a purpose in mind, typically to search for something.
However, images must be assessed before they are used in a diagnosis to make sure they provide information that is visually clear and relevant to the specific task.
The NIH grant provided the team with funding for the research lab so they could advance their research. With this funding, Anastasio will be able to hire doctoral students and postdoctoral researchers for the project.
Get The Daily Illini in your inbox!
Once the new tools are developed, they will be made open source, Anastasio said. This means other researchers will be able to download the developed algorithms and computational tools to apply them to their specific problems.
“I expect our work will have a big impact on the field of imaging science because people now will have a new capacity for assessing image quality in a way that’s important and meaningful for medical imaging,” Anastasio noted.
Anastasio explained that the tools they are developing for assessing imaging quality could attract interest from the Food and Drug Administration.
According to FDA regulation, all commercial medical imaging technologies must be approved before it is purchased and used in a hospital. The researchers’ algorithm has the potential to make the FDA’s regulatory process much more efficient.
“We’ve really gotten a lot of traction on it and I think we’ve emerged as one of the leading groups in the world in this area,” Anastasio said.
