December 02, 2005
While the first digital, filmless hospital was created just seven years ago, PACS is now an accepted technology, offering radiologists and other clinicians the ability to retrieve, share, and remotely access complex, two-dimensional scan data. The amount of information that must be processed, however, has grown exponentially, taxing the ability of standard desktop workstations to process and display the data in a timely fashion.
Radiologists today are experiencing "slice overload;" it is simply not possible to efficiently view thousands of single images in a reasonable amount of time. As a result, more clinicians are now looking toward volume reconstructions, rather than a multitude of static scans, as an efficient and optimal use of the entire scanned data set. With volume exploration capabilities, researchers can examine scans much more accurately and faster, discovering anomalies that would simply not be apparent through a multitude of two-dimensional views.
"We're undergoing a revolution in CT scanning as a digital input modality," said Robert Cooke, Fuji's executive director of marketing, network systems. "Volume exploration will make much more accurate diagnoses possible, creating great benefits for science and ultimately the patient."
Fuji and SGI are creating a system that eliminates the digital bottlenecks of expanding data, making rapid, economical volume exploration a reality. To accomplish this, data no longer will be rendered on a radiologist or doctor's desktop but rather at a centralized 3D graphics server. Using SGI's shared-memory and single system image architecture integrated with multiple ATI FireGL graphics processor units (GPUs), data processing tasks can be divided between GPUs to minimize rendering time and maximize image quality.
"We are proud of the impact that ATI FireGL workstation graphics accelerators are having today in powering the next generation of medical imaging solutions," said Dinesh Sharma, director of Workstation products, ATI Technologies. "Together with Fuji and SGI, ATI is pleased to be playing a key role in bringing to the medical community new capabilities that will dramatically improve the diagnostic process for medical professionals and their patients."
The Silicon Graphics Prism visualization system brings the following benefits to the new Fuji volume exploration PACS solution:
-- Short response time. Radiologists and clinicians can start interacting
with the data within seconds because rendering is local to where these
large data sets are stored. SGI Visual Area Networking (VAN) technology
sends just the reconstructed voxels to the radiologist's desktop while
data still resides on the server.
-- Scalability in data handling capability. Large data sets can be loaded
in main memory due to the 64-bit architecture. System resources such as
CPUs, I/O, memory, storage, and graphics, can be independently expanded
as the hospital's needs grow.
-- Using dynamic load balancing, the radiologist is not limited to the
texture capacity of a GPU and can utilize the scalability of the
Silicon Graphics Prism architecture to load more studies for the best
diagnosis. Dynamic load balancing does not limit radiological studies
to the capacity of modern day GPUs.
-- Scalability in number of users. Multiple users can share the same
system due to scalable architecture.
-- Scalability in rendering quality. Modern GPUs coupled with scalability
allow high-quality rendering algorithms to be deployed. This way a
volume can be interactively rendered by a user and "tumbled" without
the resolution sacrifice all other 3D systems make. This is a key
enabling technology so that diagnoses can be made with volumes when
there is no a priori knowledge of where the radiologist needs to look
for potential disease processes.
-- Maintains existing workflow. The FUJIFILM and SGI solution maintains
the standard diagnostic Synapse workflow. Scans are accessed through
the standard Internet Explorer web browser interface, and data is
transferred through the IP network. Physicians use their existing
monitors and drives; no equipment upgrade or system re-education is
necessary. And because the Silicon Graphics Prism system will reside
alongside Synapse, the two will integrate seamlessly. Researchers can
access volume data using the same workflow techniques with which they
are already familiar.
The Silicon Graphics Prism system brings Visual Area Networking (VAN) technology to diagnostic scan analysis, making it possible to rapidly render and transmit volume exploration data to virtually any desktop workstation without sending the data across the network.
VAN through SGI OpenGL Vizserver software enables the transfer of rich data between the Silicon Graphics Prism and a thin client. To keep the files small, VAN technology transmits only the pixels of the rendered graphic, rather than the raw data itself. As a result, VAN technology can operate on virtually any type of client, including laptops, workstations and, eventually, even PDAs.
"The combination of Silicon Graphics Prism visualization system and Fuji's Synapse PACS solution provides hospitals with a very cost-effective, powerful and flexible centralized system. As technology grows customers will be able to leverage the latest innovations in compute and visualization without changing the entire PACS infrastructure," said Afshad Mistri, senior manager of Advanced Visualization, SGI.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?