|
Digital X-ray
"Once a patient's mammograms are loaded into the system, they can be evaluated with powerful tools that isolate abnormalities very quickly by comparing current X-rays with those from previous years. Traditional film X-rays of individual patients are often scattered among various medical facilities, making them hard to find when needed. This Grid will help ensure that all of a patient's vital data is provided to authorized physicians very quickly, efficiently and securely."
-- Dr. Robert Hollebeek, Ph.D. Director, University of Pennsylvania's National Scalable Cluster Lab
What is the Challenge? (The Need)
Just about everything is going digital today, and it was only a matter of time before technology was able to produce high enough definition to replace the standard X-ray film. General Electric has recently come out with a device intended for mammograms that uses a large, 7 x 9-in. silicon detector to transfer X-ray energy directly into electrical signals that offers 25 - 100 µm resolution - a method similar to a digital camera that takes the picture without the film and stores the images electronically. A typical mammogram incorporates four images of 8 to 10 MB per image, resulting in 40-MB digital images for a complete file. With the images in digital form, electronic manipulation becomes possible.
In the US, a GRID network has been set up to store mammograms and allow sophisticated image comparison software to check for changes from previous images - no matter where those images are actually stored. This distributed system is expected to see traffic of 28 terabytes (that's 28,672 GB or 477 x 60GB hardrives such as those found in today's PCs) worth of image data every day (http://www-3.ibm.com/e-business/doc/content/growingsuccess/univofpa.html). Luckily, the architecture of the GRID network (http://www.globus.org) allows this data to be stored locally and accessed nationally, helping to distribute the access and computation power across thousands of computers. This system allows images to be viewed by the primary physician, a geographically remote "expert" physician, or a computer-aided diagnosis system in an effort to provide the most accurate reading.
However, digital x-rays and other imaging technologies will require high bandwidth capabilities even without such a sophisticated GRID-based network in place. In Ballarat, Victoria, the Ballarat Base Hospital has a digital X-ray centre that is the largest of its kind in Australia. For sophisticated 3D images of the human body, technicians were taking up to 100 images that would be merged into a 3D model - and that model could take up to half a gigabyte of data. The only solution they had was to burn the data onto a couple of compact disks and send those with the patient to the ward (http://computerworld.co.nz/) - the existing network speed would take hours, if not days to transmit the information. There has to be a better way!
How Can NGI Help? (The Use)
The Ballarat Base Hospital installed a Gigabit Ethernet with points of presence throughout the hospital complex. They did this with financial help from the Collaborative Optical Leading Testbed (COLT) - COLT is accelerating the deployment of Next Generation Internet communications in regional Victoria resulting in millions of dollars of exports and the creation of many new jobs. The network plan is to connect the local hospitals and universities, and then provide connectivity, which will pay for the incremental upgrades to provide for local business and from there to small businesses and eventually residential users.
With Internet speeds 50 to 100 times faster than dial-up, single medical images can now be transferred in 20 seconds instead of 30 minutes. This allows Doctors throughout the hospital to pull up images as needed and share them with other Physicians that are geographically remote - even if it they are only in a different building.
Pilot projects such as Australia's COLT system demonstrate the value of high speed Internet on a small scale, while larger projects like the National Digital Mammography Archive (NDMA) [US] and the UK's eDiamond (a British Mammography archive and diagnosis system) demonstrate the abilities of a system once all of the nodes in the network have been connected. With annual NDMA volumes in excess of 5.6 petabytes per year - even NGI will be stretched!
The Bottom Line (The Value)
The NDMA and the eDiamond project were created due to some alarming statics. Experts estimate that one fifth of first-time mammograms are misdiagnosed. And a separate, but equally disturbing problem is that in 20 percent of cases, previously prepared images cannot be located for comparison purposes (The Grid: Computing without Bounds, Scientific American).
These digital archives are addressing these problems through distributed and secure storage (which is accessible by Physicians anywhere, anytime), high-speed connectivity to move the huge amount of data contained in the images, and access to a diagnostic program that compares images to the archive's records (similar to finger print identification) to catch cancer early enough for treatment. The above projects also are expected to help reduce the rate of false-positive diagnosis, overcome the challenge of inconsistent image formats and lost films which prevent proper diagnosis, while also allowing physicians to study and compare similar cases so they can develop better treatment options.
In 2002, the US calculated the total national economic cost of cancer to be US$171.6 billion, with US$6 billion (www.cancer.org/docroot/MIT/content/MIT_3_2X_Costs_of_Cancer.asp) related directly to breast cancer. The burden of cancer on New Zealand is even greater (cancer deaths are 9% higher than in the US, and 3% higher than the UK per capita [http://web.hhs.se/personal/suzuki/o-English/he02.html]), bringing the need for better tools that much closer to home.
|
|
