Mercury Computer Systems, Inc. (NASDAQ: MRCY) announced a cooperation with the Institute of Medical Physics (IMP) of Erlangen, Germany, which is focused on the joint development and commercialization of medical imaging technology deployed on the Cell Broadband Engine ™ (BE) processor.
Under the terms of the agreement, Mercury is working together with the IMP on designing and implementing ambitious reconstruction and visualization algorithms with real-time performance on the Cell BE processor, to deliver orders-of-magnitude performance increases while also reducing the complexity and costs of medical image processing systems. Together with IMP technology, Mercury will integrate the algorithms into high-performance Cell BE processor-based systems designed to significantly accelerate the reconstruction and visualization of medical imaging data for medical OEMs.
Medical image reconstruction, such as that found in CT, tomosynthesis, PET and SPECT, is computationally demanding. Advances in medical sensor technology have created an increasing number of images per procedure, posing a tremendous challenge for processing and visualization in a timely manner that is compatible with the hospital workflow. As example of the joint work, Mercury and the IMP have developed a Cell BE processor-based solution capable of performing modern CT reconstruction more than 100 times faster than conventional microprocessors. The level of parallelism along with the vast I/O capabilities permits the Cell BE processor to efficiently implement complex CT reconstruction algorithms with close to real-time performance. The Cell BE processor enables system design in which the radiologist can view images obtained from better algorithms, with higher quality, much sooner than ever before.
“With Cell Technology, advanced approaches thought to be too demanding in terms of processing can now enter daily routine,” said Marc Kachelriess, Professor of Medical Imaging at the IMP. “The tremendous processing power of the Cell BE processor enables the use of iterative reconstruction algorithms designed for the reduction of beam hardening and metal artifacts. Analytical and statistical CT reconstruction algorithms that allow for enhanced image quality, while keeping the X-ray exposure of the patient as low as possible, can also be implemented on a Cell BE processor-based platform.”
The Cell BE processor, developed by IBM, Toshiba and Sony, is a breakthrough architectural design featuring eight synergistic processing elements that provides unmatched performance levels in many computationally intense applications. The Cell BE processor has peak performance in excess of 180 GFLOPS − which equates to 180 billion floating-point operations per second and an amazing 25 GB/s memory bandwidth. Mercury has a multi-year agreement with IBM to deploy the revolutionary Cell BE processor in computer systems for defense, life sciences, seismic, and industrial applications. Mercury is offering this technology to medical OEMs worldwide via the Cell Acceleration Board (CAB) – a PCI Express® accelerator card based on the Cell BE processor – along with algorithms, and multicore and multicomputer programming expertise.
“With Cell BE processor technology, Mercury offers its medical OEM customers the advantages of a cost-effective, high-performance platform for reconstruction and imaging” said Marcelo Lima, Vice President and General Manager of Commercial Imaging and Visualization at Mercury. “In addition, we can offer our customers access to some of the most advanced algorithm know-how on the planet, both through the IMP and Mercury, and help our customers get this advanced technology to market faster.”
For more information on Mercury’s advanced solutions for medical imaging, visit Mercury at RSNA 2006 November 26-30 at McCormick Place in Chicago, Illinois, in the South Building – Hall A, Booth #4383; visit www.mc.com/RSNA06; or contact Mercury at (866) 627-6951.
About the Institute of Medical Physics (IMP) of Erlangen (Germany)
The development and the application of imaging technologies and modalities for medical diagnosis, therapy planning and computer assisted surgery represent a major focus of IMP’s research activities. In particular, projects related to X-ray computed tomography (CT), among them the development of cardiac CT, are recognized worldwide and allowed the IMP to reach a leading position internationally. Basic research activities of the IMP relate to detector technologies for x-ray diagnosis which are of interest to medical diagnosis as well as to non-destructive materials testing. For this purpose a universal CT setup and a micro-CT apparatus have been developed which allow scanning in cone-beam geometry with a flat panel detector of 1024x1024 detector elements and more. A high-end clinical spiral cone-beam CT scanner as well as C-arm CT scanners are available for basic research and advanced clinical studies. Magnetic resonance tomography, ultrasound and positron emission tomography supplement the spectrum of activities. Topics related to image quality, to quality assurance and to patient dose assessment and patient dose reduction are pursued with respect to all these activities.
For more information please visit www.imp.uni-erlangen.de.
Under the terms of the agreement, Mercury is working together with the IMP on designing and implementing ambitious reconstruction and visualization algorithms with real-time performance on the Cell BE processor, to deliver orders-of-magnitude performance increases while also reducing the complexity and costs of medical image processing systems. Together with IMP technology, Mercury will integrate the algorithms into high-performance Cell BE processor-based systems designed to significantly accelerate the reconstruction and visualization of medical imaging data for medical OEMs.
Medical image reconstruction, such as that found in CT, tomosynthesis, PET and SPECT, is computationally demanding. Advances in medical sensor technology have created an increasing number of images per procedure, posing a tremendous challenge for processing and visualization in a timely manner that is compatible with the hospital workflow. As example of the joint work, Mercury and the IMP have developed a Cell BE processor-based solution capable of performing modern CT reconstruction more than 100 times faster than conventional microprocessors. The level of parallelism along with the vast I/O capabilities permits the Cell BE processor to efficiently implement complex CT reconstruction algorithms with close to real-time performance. The Cell BE processor enables system design in which the radiologist can view images obtained from better algorithms, with higher quality, much sooner than ever before.
“With Cell Technology, advanced approaches thought to be too demanding in terms of processing can now enter daily routine,” said Marc Kachelriess, Professor of Medical Imaging at the IMP. “The tremendous processing power of the Cell BE processor enables the use of iterative reconstruction algorithms designed for the reduction of beam hardening and metal artifacts. Analytical and statistical CT reconstruction algorithms that allow for enhanced image quality, while keeping the X-ray exposure of the patient as low as possible, can also be implemented on a Cell BE processor-based platform.”
The Cell BE processor, developed by IBM, Toshiba and Sony, is a breakthrough architectural design featuring eight synergistic processing elements that provides unmatched performance levels in many computationally intense applications. The Cell BE processor has peak performance in excess of 180 GFLOPS − which equates to 180 billion floating-point operations per second and an amazing 25 GB/s memory bandwidth. Mercury has a multi-year agreement with IBM to deploy the revolutionary Cell BE processor in computer systems for defense, life sciences, seismic, and industrial applications. Mercury is offering this technology to medical OEMs worldwide via the Cell Acceleration Board (CAB) – a PCI Express® accelerator card based on the Cell BE processor – along with algorithms, and multicore and multicomputer programming expertise.
“With Cell BE processor technology, Mercury offers its medical OEM customers the advantages of a cost-effective, high-performance platform for reconstruction and imaging” said Marcelo Lima, Vice President and General Manager of Commercial Imaging and Visualization at Mercury. “In addition, we can offer our customers access to some of the most advanced algorithm know-how on the planet, both through the IMP and Mercury, and help our customers get this advanced technology to market faster.”
For more information on Mercury’s advanced solutions for medical imaging, visit Mercury at RSNA 2006 November 26-30 at McCormick Place in Chicago, Illinois, in the South Building – Hall A, Booth #4383; visit www.mc.com/RSNA06; or contact Mercury at (866) 627-6951.
About the Institute of Medical Physics (IMP) of Erlangen (Germany)
The development and the application of imaging technologies and modalities for medical diagnosis, therapy planning and computer assisted surgery represent a major focus of IMP’s research activities. In particular, projects related to X-ray computed tomography (CT), among them the development of cardiac CT, are recognized worldwide and allowed the IMP to reach a leading position internationally. Basic research activities of the IMP relate to detector technologies for x-ray diagnosis which are of interest to medical diagnosis as well as to non-destructive materials testing. For this purpose a universal CT setup and a micro-CT apparatus have been developed which allow scanning in cone-beam geometry with a flat panel detector of 1024x1024 detector elements and more. A high-end clinical spiral cone-beam CT scanner as well as C-arm CT scanners are available for basic research and advanced clinical studies. Magnetic resonance tomography, ultrasound and positron emission tomography supplement the spectrum of activities. Topics related to image quality, to quality assurance and to patient dose assessment and patient dose reduction are pursued with respect to all these activities.
For more information please visit www.imp.uni-erlangen.de.
