The Mayo Clinic and IBM have partnered in a venture to improve medical imaging technology. The clinic's current technologies aren't keeping up with the intense processing demands required to analyze digital medical images such as x-rays, CT scans, and MRIs. You've learned in this chapter that transistor densities on a chip double roughly every two years, a rule of thumb referred to as Moore's Law. Bradley Erickson, chairman of radiology at the Rochester-based Mayo Clinic, was quoted in Computerworld as saying, "We are facing significant problems in medical imaging because the number of images produced in CT scanners basically tracks Moore's Law. My eyes and brain can't keep up. I see more and more images I have to interpret. ... The innovation here is to take computer chips and extract the information in these increasing number of images and help present it usefully to the radiologist." This is a case of technology outpacing the human ability to manage the information it produces. In such cases, we turn to technology for solutions. For doctors and radiologists at the Mayo Clinic, standard computer processors cannot keep up with their need to analyze digital images. So they are turning to the Cell processor from IBM in hopes that it will provide a solution. The Cell processor is the chip that makes Sony's PlayStation video-game console the most powerful console in the industry, according to many game enthusiasts. The Cell processor was created in a joint effort by IBM, Sony, and Toshiba, with an architecture that is specially designed to accelerate graphics processing. Researchers at IBM and Mayo believe that it could turn a 10-minute CT image analysis into a four-second job. One of the tasks in which the Cell processor could be useful is in comparing scan images of a patient over time. For example, to track the progression or regression of cancer in a patient, physicians compare CT scans of the tumor over time to look for change. Changes are often too subtle for the human eye to notice, so software that implements a complex algorithm is used to analyze the photos. Using a standard PC processor, the algorithm may take several minutes to complete. While this may not sound like much, typically a physician needs to run several analyses in sequence, consuming significant amounts of time. The process of transforming 2-D images into 3-D-something the Cell processor was designed for-also requires significant time using traditional processors. With the Cell processor, these tasks might be completed in a matter of a seconds. Mayo Clinic and its effort to speed up the analysis of image scanning illustrate the importance of time when it comes to processing. Whether it's working to save a life, to finish design specifications for a new product, or to analyze stock market trends, the difference between a minute and a second can mean success or failure. For professionals in most industries, having the best processor for the task at hand, and matching it with the best hardware and software, provides them with a winning solution.
Discussion Questions
1. Why is the Cell processor the best processor for the Mayo Clinic's tasks? How might it empower physicians to save more lives?
2. If the Cell processor is so much faster than typical PC processors, why isn't it being used in PCs?