Demands on information systems to manipulate 2D, 3D, and 4D images, provide video conferencing, and satisfy other evolving telemedical applications require advanced networking solutions. Grid computing has emerged as one answer.
Demands on information systems to manipulate 2D, 3D, and 4D images, provide video conferencing, and satisfy other evolving telemedical applications require advanced networking solutions. Grid computing has emerged as one answer.A recent paper from Greece proposes a data-centric network that employs Grid architecture to construct web PACS portals to meet contemporary radiology department data management needs (Conf Proc IEEE Eng Med Biol Soc 2007;1:6429-33). Grid computing is a hardware and software infrastructure capable of providing reliable, low-cost access to high-end computing."This architecture demonstrates greater computing and data storage capabilities by taking advantage of many networked computers operating as one, distributing process execution via parallel infrastructure," said George Koutelakis, Ph.D., an electrical and computer engineer at the University of Patras.The proposed architecture yields a large-scale computing network from inexpensive off-the-shelf servers and storage devices. All components perform as a logical portal platform for providing web PACS services, Koutelakis said."Each section of a hospital radiology department does not need autonomous computing power to process images," Koutelakis said. "Rather, each can obtain computing resources through an application server on the Grid." Provision of different services through distributed distinguished servers reduces the load that each server manages. In this way, PACS servers achieve service multitasking. Servers that accept images from modalities (Store requests), for example, are distinguished from servers that provides Query/Retrieve services to terminal users. Servers providing the aggregate of PACS services constitute a cluster of servers. Web PACS applications are developed by employing distributed/parallel algorithms. Thus, a user connected to the network can receive computing power, storage space, and other resources from other systems on the Grid. Requests are filtered through a virtualization layer, which identifies and allocates necessary resources. Assignments are based on priority and service level agreements.
Implementation requires three sets of applications:
"Grid PACS is an effort to develop PACS services over a Grid platform to allow radiologists and clinicians to respond to new data management challenges efficiently," Koutelakis said.
The Grid solution could conceivably be expanded to the entire hospital, although a thorough enterprise cost-benefit analysis would first be required, he said.
CT Study Reveals Key Indicators for Angiolymphatic Invasion in Non-Small Cell Lung Cancer
January 15th 2025In computed tomography (CT) scans for patients with solid non-small cell lung cancer (NSCLC) < 30 mm, emerging research suggests the lollipop sign is associated with a greater than fourfold likelihood of angiolymphatic invasion.
Can MRI-Based AI Enhance Risk Stratification in Prostate Cancer?
January 13th 2025Employing baseline MRI and clinical data, an emerging deep learning model was 32 percent more likely to predict the progression of low-risk prostate cancer (PCa) to clinically significant prostate cancer (csPCa), according to new research.
Shaping the Future of Radiology in 2025: Trends, Threats, and Opportunities
January 10th 2025How do we respond to challenges with staff recruitment, cybersecurity, and looming hospital takeovers in radiology? This author assesses key trends in radiology and offers key insights to stay competitive in the field.