backtop


Print 33 comment(s) - last by teainthesahara.. on Apr 15 at 6:33 AM


Cell Broadband Engine becoming a medical powerhouse
The PS3 wonderchip does more than play games -- it saves lives

The Cell Broadband Engine is a truly versatile piece of silicon. It’s inside every PlayStation 3 powering games, decoding Blu-ray movies and curing diseases with Folding@home. It’s also inside a few IBM BladeCenter servers, which will soon be utilized for medical imaging. Collaborators from Mayo Clinic and IBM say that they are now using the Cell Broadband Engine to dramatically speed up the processing of 3D medical images.

The advance significantly aids image registration -- the computer-enhanced alignment of two medical images in three-dimensional space. One way medical images are being improved is by using visual images from more than one source -- magnetic resonance imaging (MRI) and computerized tomography (CT) scans for example. With the images properly aligned over one another, a radiologist can more easily detect structural changes such as the growth or shrinkage of tumors.

"This alignment of images both improves the accuracy of interpretation and improves radiologist efficiency, particularly for diseases like cancer," says Mayo radiology researcher Bradley Erickson, M.D., Ph.D.

Through porting and optimization of Mayo Clinic's Image Registration Application on the IBM 'Cell Blade,' the application produces image results fifty times faster than the application running on a traditional processor configuration. Mayo Clinic and IBM used 98 sets of images and ran the optimized registration application on the IBM BladeCenter QS20 with the Cell Broadband Engine, in comparison with running the original application on a typical processor configuration. The application running on a typical processor configuration completed the registration of all 98 sets of images in approximately 7 hours. The team adapted the  application optimized for Cell and completed the registration for all 98 sets of images in just 516 seconds, with no registration taking more than 20 seconds.

"This is all about taking technology innovation, collaborating with our customers, and applying it to help them directly benefit their patients," said Shahrokh Daijavad, Next Generation Computing, Systems & 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."

Sony also revealed plans to lease out the computing power of the Cell Broadband Engine inside PlayStation 3 consoles as part of a supercomputer grid.



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

By saratoga on 4/14/2007 2:35:01 PM , Rating: 2
quote:
A couple of years ago I did software for a company that did digital xray imaging (specifically digital mammography). We could do and display realtime (or near realtime) digital xrays, but one of the things that took compute power was doing enhancement of the image. I convinced them to move from a sun blade to a dual opteron which cut the time to a third of previous. I knew stuff like the cell could really help and would probably be able to do it in close to realtime which would have been great. Unfortunately, the company went bellyup.


These days FPGA systems are getting popular. For that sort of thing, some of the new 3D imaging systems being demoed have basically a U1 sized board that streams data from the ADC channels and implements all of the forming/DSP/etc stuff in hardware. Literally, an adder, a register, multiply by a coefficient etc to implement an FIR filter or whatever. Really cool idea, but inflexible and expensive.

I think chips like Cell could revolutionize how some of these products were designed. FPGAs and dual Xeons are used because there were no DSP cores that could handle the bandwidth and FLOPS needed to do high FPS and high resolution in real time. But with something like Cell you can, all in one chip, and for a tiny fraction of the price. Not to mention the size savings. Used to be these systems were the size of a small rack. The new FPGA systems are almost PC sized. The next gen will probably be the size an Xbox 360 :)

Which is awesome, I hate moving test systems between buildings.

Also the cool thing about this is the possibilities opened up by having all those FMACs/clock. Right now manufacturers quietly implement all sorts of "optimized" algorithms to avoid having to do large (but more accurate and more useful) cross correlations. But on a chip like Cell, you could easily do a cross correlation across a huge dataset, in real time, and using only a few SPEs. Off the top of my head I can think of a few really cool motion tracking applications, and I'm sure theres tons more that the engineers have sitting in the lab.


"Well, there may be a reason why they call them 'Mac' trucks! Windows machines will not be trucks." -- Microsoft CEO Steve Ballmer




Latest Headlines
4/21/2014 Hardware Reviews
April 21, 2014, 12:46 PM
4/16/2014 Hardware Reviews
April 16, 2014, 9:01 AM
4/15/2014 Hardware Reviews
April 15, 2014, 11:30 AM
4/11/2014 Hardware Reviews
April 11, 2014, 11:03 AM










botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki