Internet2/NLM
infoRAD Demos and Tutorials at RSNA
Around 30,000 radiologists, healthcare administrators, technologists
and scientists, plus another 30,000 imaging and information system
vendors from all over the world, are expected to attend the Radiological
Society of North America (RSNA) annual meeting, December 1-6, in
Chicago, making it the world's largest medical meeting. RSNA supports
the exchange of scientific progress in radiology, radiologic education,
and in the integration of information and communication systems
in radiology practice. The Annual Meeting provides workshops, training,
and conference sessions for health professionals. In addition, nearly
500,000 square feet of exhibit space – feature product demonstrations
and a broad range of exhibits. The Internet2/NLM tutorials and demos
will be part of the infoRAD exhibit space. The info RAD area is
designed to showcase the most innovative technology solutions in
an interactive, educational environment. Hands-on demonstrations
are encouraged. Computer-aided instruction, digital imaging and
communications in medicine, new technologies, computer-assisted
diagnosis, and literature searches are just some of the topics featured
in infoRAD exhibits.
Internet2/NLM infoRAD Tutorial Schedule
|
|
|
Monday, December 2
|
10:30 a.m. – 12:00 p.m.
1:30 – 3:00 p.m. |
|
Tuesday, December 3
|
10:30 a.m. – 12:00 p.m.
1:30 – 3:00 p.m. |
|
Wednesday, December 4
|
10:30 a.m. – 12:00 p.m.
1:30 – 3:00 p.m. |
|
Thursday, December 5
|
10:30 a.m. – 12:00 p.m.
1:30 – 3:00 p.m. |
|
Featured Demos for infoRAD
The following demos will be featured at the Internet2/NLM tutorials
in the infoRAD exhibit area at RSNA.
Internet2/NLM infoRAD Demos in the News
Log
in to the Future at RSNA 2002, RSNA News, August 2002
Advanced Biomedical
Tele-Collaboration
http://www.accessgrid.org/
Developed at:
University of Chicago
Demonstrators:
Jonathan Silverstein
Fred Dech
Contact:
Jonathan Silverstein
jcs@uchicago.edu
(773) 834-8200
Description:
Biomedical research, education and medical practice have become
socially complex activities. Coordination of complex sources of
data, information, knowledge and experience is now critical. Our
laboratory focuses on prototyping applications which enable quality
collaborations in biomedicine. By combining teleconferencing and
virtual reality in self-scaling, remotely controlled, distributed
computing architectures, advanced tele-collaboration permits coordination
among colleagues in real-time. We combine group to group interaction
technologies with scientific visualization creating new environments
that permit physicians to see each other's point of view and to
enhance these with one's own experience. We will demonstrate shared
audio, video and immersive virtual reality through a modified Access
Grid Node. This highlights the infrastructure on which remotely
controlled, high performance and wearable collaborative computing
architectures will be enabled.
Role of Internet2:
Tele-Immersion, or real-time collaboration of remotely connected
sites in a Virtual Reality environment, is essential to the future
investigation of radiological data. The difficulties and variables
involved in developing a successful Tele-Immersive application are
many. Shared or synchronized databases, application hand-shaking,
streaming audio and video, mobile security/encryption and multicast
network management are issues that require detailed attention. A
shared distributed virtual environment enables us to realize real-time,
immersive teleradiology. The Access Grid is the ensemble of resources
that is used to support human interaction across the grid. It consists
of multimedia display, presentation and interactive software environments,
interfaces to grid middleware, and interfaces to visualization environments.
The Access Grid technology was originally developed by the Futures
Laboratory in the Mathematics and Computer Science Division at Argonne
National Laboratory.
Internet2 Performance for
Medical Imaging Applications
ftp://128.125.76.37/pub/download/IPI2002Report.pdf
Developed at:
Childrens Hospital Los Angeles
University of Southern California
Demonstrators:
Greg Mogel
Fei Cao
H. K. Huang
Michael Zhou
Brent Liu
Cammy Huang
Contact:
Fei Cao
fcao@pacbell.net
(323) 671-3848
Partners:
St. John's Hospital
PACS Lab, UCLA Medical Center
Virtual Labs, Stanford University Medical School
Cedara Software Corp.
Shanghai Institute of Technical Physics
Description:
IPI Laboratory (Image Processing and Informatics) at Childrens
Hospital Los Angeles and USC was a pioneer in using the Internet2
for medical imaging applications, through the Next Generation Internet
supports from the National Library of Medicine and the TATRC, Army
Medical Research and Materiel Command since 1999. At the IPI Lab,
we have a dedicated OC-3 line connected via the University of Southern
California to the CalREN2 GigaPoP of the Internet2. In the past
several years, we have accumulated Internet2 performance data in
medical imaging applications through several Internet2 connections
including UCLA, UCSF, Stanford University, University of Hawaii,
and the National Library of Medicine.
In this demonstration, we will set up two 2,000 line LCD display
workstations at the demo site with Internet2 connection to the PACS
(Picture Archiving and Communication System) Simulator Server at
the IPI Lab. Three medical image applications with hands-on demos
will be given:
1. PACS ASP (Application Service Provider) image off-site back-up
archive for Disaster Recovery.
2. PACS DICOM (digital image communication of medicine) remote
image query, retrieval and display.
3. Innovative technology in interactive teaching of biomedical
curriculum via the Web.
Role of Internet2:
Medical image data is different from other types of data because
the file sizes are large (hundreds of Mbytes per exam) and the required
turn-around time is short. The high bandwidth and low latency of
Internet2 have made it feasible to transfer large quantitues of
medical image data over low cost and high-speed wide-area networks,
especially for our applications in off-site backup, teleradiology,
and remote interactive teaching.
Multi-Center Clinical
Trial Using NGI
http://www.isis.georgetown.edu/ald/
Developed at:
Kennedy Krieger Institute
Demonstrators:
Hugo Moser
Betty Levine
Sayed Ali Fatemi
Mary Lou Ingeholm
Contact:
Hugo Moser
moser@kennedykrieger.org
(443) 923-2750
Betty Levine
levineb@georgetown.edu
(202) 687-7950
Partner:
Georgetown University
Funded by:
NLM Next Generation Internet Initiative
Description:
This National Library of Medicine funded project uses the Internet2
as a mechanism for facilitating multi-center clinical trials for
a rare disease, X-linked adrenoleukodystrophy (X-ALD). Using a digital
imaging network and Internet2 high-performance networks allows researchers
to share medical images and interpretation of results from multiple
institutions efficiently.
X-ALD is a rare neurological disorder affecting 1:17,000 people
worldwide. A brain MRI permits early detection of nervous system
damage. The Loes severity score (0-34 points) is used for clinical
evaluation of X-ALD MRI images and is based on location and extent
of disease and the presence of focal and/or global atrophy. It is
used to monitor disease progression and in determining therapy efficacy.
The digital MRI network permits researchers to share MRI images
and resultant Loes scores from multiple institutions efficiently,
providing an infrastructure for conducting multi-center clinical
trials.
Role of Internet2:
The X-ALD MRI network uses the DICOM 3.0 standard to move the MRI
image sets securely over the commodity Internet and Internet2 networks
to a secure central clinical database. Virtual private network (VPN)
technology is used to protect patient confidentiality and assure
data integrity. The network operates in the standard Internet and
Internet2 environments, providing the opportunity to evaluate and
compare the impact of near real-time access to the MRI studies.
Internet2 connectivity improves access to the large image sets and
thus improves the ability for real-time collaboration and faster
clinical evaluation. Internet2 access permits the image set transfer
speeds required for near real-time teleradiology for multi-center
clinical trials.
Multi-Site 3-D Tomosynthesis
Mammography: Image Acquisition, Reconstruction, CAD and Display
Developed at:
Uniformed Services University of the Health Sciences
Demonstrators:
Jerry Thomas
Jeff Eberhard
Maria Kallergi
Contact:
Jerry Thomas
jthomas@usuhs.mil
(301) 295-3246
Partners:
General Electric Global Research
University of South Florida
Description:
Telemammography of 3D tomosynthesis images will be demonstrated.
Images will be acquired at National Naval Medical Center, transferred
to the Uniformed Services University of the Health Sciences (USUHS)
for storage. General Electric Global Research will apply advanced
reconstruction techniques and volume rendering processing to the
image data sets. Slice and volumetric data will be processed with
3D CAD algorithms at University of South Florida. All processed
data will be returned to USUHS where study interpretation will be
performed. This demonstration will replace the USUHS interpretation
"corner" of the data triangle.
Role of Internet2:
Internet2 high-performance networks will be used to transfer raw
projection data (100 Mbytes/breast, 200 Mbytes/patient) to a remote
site for reconstruction into planar slices (between 320 and 650
Mbytes/breast, or 640 Mbytes to 1.3 Gbytes/patient) and volumetric
renderings (500 to 800 Mbytes/rendering/breast, 1Gbyte to 1.6 Gbytes/patient).
Reconstructed data will be returned to the acquisition/storage site
where it will be accessed by a CAD development site for CAD processing.
Planar reconstructions, volumetric renderings and CAD results will
be displayed at the acquisition site for clinical interpretation.
Current methods require data to written to CD and DVD and mailed
to the remote sites. Internet2 connectivity will increase the research
productivity and provide rapid feedback which will improve the reconstruction,
volume rendering and CAD research efforts.
National Digital Mammography Archive:
Clinical and Research Components
http://nscp01.physics.upenn.edu/ndma/
Developed at:
University of Pennsylvania
Demonstrators:
TBD
Contact:
Mitchell D. Schnall
schnall@oasis.rad.upenn.edu
(215) 662-7238
Barbara
G. Beckerman
beckermanbg@y12.doe.gov
(865) 576-2681
Partners:
BWXT Y12
University of North Carolina at Chapel Hill
University of Chicago
University of Toronto
LAItechnologies
Description:
The NDMA represents a collaborative effort between the teams
mentioned above to develop a testbed that demonstrates the feasibility
of an integrated, scalable, secure national breast imaging archive
and network infrastructure to support digital mammography using
advanced network technologies. Applications are being developed,
tested and integrated by all the participating team members. The
team is leveraging a high speed network infrastructure to improve
access and performance of breast cancer screening with an imaging
archive that supports storage, retrieval and distribution of breast
images and reports for clinical and research purposes and ensures
privacy with multi-level security embedded throughout the system.
NGI and /or Internet 2 technologies are need to schedule transfer
of large data files, execute real-time queries, and access information
securely. The applications also permit secure sharing of patient
images and related reports across healthcare enterprises.
Role of Internet2:
Multi-site data intensive environment for healthcare records uses
encrypted VPNs utilizing Internet2 resources connecting CaNet3,
Abilene, and ESNet. This project benefits particularly from the
high bandwidth available to the multi-sites for Internet2. Projects
dealing with digital mammograms—which are extremely large, uncompressed
images ranging in size from 8 MB to 50 MB each—require high bandwidth.
These projects also benefit from low latency of the network. The
network sockets and applications have been expressly tuned to provide
good performance on the high-delay bandwidth networks.
Visuo-Haptic Applications
for Anatomy and Surgery Education Over the NGI
http://ngi.stanford.edu/
Developed at:
Stanford University
Demonstrators:
Parvati Dev
Steven Senger
Dale Harris
W. LeRoy Heinrichs
Contact:
Parvati Dev
parvati@stanford.edu
(650) 723-8087
Partner:
University of Wisconsin, La Crosse
Funded by:
NLM Next Generation Internet Initiative
Description:
We will demonstrate several applications being developed under
a National Library of Medicine NGI contract. These will include
the Remote Stereo Viewer, Immersive Segmentation and CEAnatomy applications.
The Remote Stereo Viewer provides on demand access to large stereo
image sets of anatomical structures. Image sets consist of a full
rotation of the structure at various levels of dissection and typically
contain several hundred high-resolution images. The Immersive Segmentation
application allows users to interactively visualize and segment
volumetric data sets such as the Visible Human using a stereoscopic
immersive interface. The system provides the user with the ability
to intuitively steer high-cost segmentation algorithms. The system
also incorporates the use of haptics to assist with the segmentation
of anatomical structures. The CEAnatomy application, implements
a client/server system for interactively navigating volumetric data
sets from a handheld PocketPC.
Role of Internet2:
The applications span a range of network bandwidth/latency requirements
and depend on the high-performance networking capabilities of Internet2.
The Remote Stereo Viewer and Immersive Segmentation applications
have burst data rates of 30 Mbps per data stream. The Remote Stereo
Viewer offers collaboration capabilities using a multicast data
stream. The Immersive Segmentation application can support multiple
visualization streams of the virtual workspace. Each stream can
be unicast or multicast. The applications which incorporate haptics
are sensitive to latency. The CEAnatomy application has high-bandwidth
requirements for a wireless application and requires low latency
to the server to support a smooth interactive interface.
|
