Domain-specific computing research looks to boost low-dose CT efficiency

A group of engineering researchers from UCLA, Rice University and the Oregon Health and Science University have received a $3 million grant from the National Science Foundation and Intel Corp. to advance medical imaging--like low-dose computed tomography--through the further development of domain-specific computing.

The majority of researchers come from the UCLA-based Center for Domain-Specific Computing (CDSC), which was awarded a five-year $10 million grant by the National Science Foundation (NSF) in 2009. The new grant comes from the Intel's and NSF's Innovation Transition program.

According to the CDSC, in domain-specific computing, researchers can create custom hardware that can solve a variety of related problems within a specific domain with high efficiency and flexibility. Consequently, that allows them to achieve a level of power and efficiency that is much greater than what can be achieved with general purpose computing.

"We tried to create energy-efficient computers that are more like brains," Jason Kong, director of CDSC, said in an announcement. "We don't really have a centralized CPU in there. If you look at the brain, you have one region responsible for speech, another region for motor control, another region for vision. Those are specialized 'accelerators.' We want to develop a system architecture of that kind, where each accelerator can deliver a hundred to a thousand times better efficiency than the standard processors."

This is one of many efforts currently underway to develop imaging techniques that can reduce patient radiation exposure. Many of these efforts are specifically aimed at reducing the radiation dose in pediatric scans.

For example, pediatric neurosurgeons and neuroradiologists at Johns Hopkins Children's Center have developed an imaging technique that can cut down on the amount of radiation exposure to children undergoing multiple brain scans. The technique uses fewer slices of the brain taken by CT scanners and can reduce radiation exposure by an average of 92 percent per patient compared with standard CT exams.

The new funding will go toward helping CDSC develop accelerator-rich architectures to be applied to medical imaging and patient-specific cancer treatments. CDSC researchers have already made a leading CT image-reconstruction algorithm 100 times faster, the development of which could significantly reduce a patient's radiation exposure.

"The low-dose CT scan allows you to get a similar resolution to the standard CT, but the patient can get several times lower radiation," Alex Bui, a professor in the radiological sciences department at UCLA and a co-leader of the project, said. "Anything we can do to lower that exposure will have a significant health impact."

To learn more:
- see the announcement from UCLA