Mayo Clinic and IBM are developing a parallel computer architecture and memory bandwidth in a joint project designed to speed the processing of 3D medical images. The technology aims at advancing image registration, which is 3D computer-enhanced alignment of two medical images obtained at different dates or by using different imaging devices.
“This alignment of images both improves the accuracy of interpretation and improves radiologist efficiency, particularly for diseases like cancer,” says Bradley Erickson, MD, PhD, radiology researcher at the Mayo Clinic.
The developers are porting and optimizing Mayo Clinic’s Image Registration Application on an IBM BladeCenter QS20 Cell Blade. Through the process the application is able to produce image results 50 times faster than the application running on a traditional processor configuration.
For this imaging project, Mayo Clinic and IBM used 98 sets of images and ran the optimized registration application on the IBM BladeCenter QS20, in comparison with running the original application on a typical processor configuration. The conventional processor completed the registration of all 98 sets in about 7 hours.
However, with an adapted “mutual-information-based” 3D linear registration algorithm application optimized for Cell/B.E. (Cell Broadband Engine) the team was able to complete the registration for all 98 sets of images in just 516 seconds, with no registration taking more than 20 seconds. The Cell/B.E. architecture is a new processor architecture which extends IBM’s 64-bit Power Architecture technology
The 3D linear algorithm works by finding the best spatial positioning to maximize the amount of information gathered from two images, thereby optimizing sampling quality while reducing sampling time.
“This is all about taking technology innovation, collaborating with our customers, and applying it to help them directly benefit their patients,” said Shahrokh Daijavad, of next generation computing, systems and technology, IBM. “This improvement with the application running on Cell, will achieve two things: allow for Mayo’s doctors and radiologists to achieve in seconds what used to take hours, which in turn will significantly decrease the wait time and anxiety for a patient waiting on news from the doctor.”
The results will be presented in a joint presentation at the IEEE (Institute of Electrical and Electronics Engineers) International Symposium on Biomedical Imaging in Washington, D.C., April 12–15.