Y. Kim, J. E. Cabral Jr., D. M. Parsons, G. L. Lipski, R. H. Kirchdoerfer*, A. Sado*, G. N. Bender* and F. Goeringer**
Department of Electrical Engineering, FT-10, University of Washington, Seattle, WA 98195
*Madigan Army Medical Center, Tacoma, WA 98431
**USAMMA - MDIS Project Office, Ft. Detrick, Frederick, MD 21702
Telemedicine is becoming increasingly possible due to the confluence of ongoing technical advances in such areas as telecommunications, imaging, multimedia, computers, and information systems. Project Seahawk is a regional telemedicine program in the Pac ific Northwest with Madigan Army Medical Center (MAMC) as the hub connecting various military and other federal hospitals and clinics utilizing the state-of-the-art technologies.
The first phase of Project Seahawk successfully connected MAMC in Tacoma, WA to the University of Washington in Seattle, WA through the Western Washington Local Access Transport Area (LATA) Integrated Optical Network (LION) Sonet Ring using Asynchronous Transfer Mode (ATM) and two MediaStation 5000s as a feasibility demonstration. Several telemedicine scenarios were demonstrated including synchronized image manipulation, real-time transmission of ultrasound and medical images, and video and audio telecon ferencing, and remote consultation. The second phase implementation will consist of increasing the number of hospitals and clinics with telemedicine capability, e.g., Bremerton Naval Hospital, Oak Harbor Naval Hospital, Seattle VA, and American Lake VA.
Project Seahawk is an effort to link federal hospitals, clinics and remote locations in the Puget Sound to enable health care providers to exercise their expertise (as if they were physically present) to the location of patients or other collaborating care providers. It will manage the limited resources more effectively and allow remote clinics to access specialty care. Project Seahawk includes Department of Defense tri-service and Department of Veteran's Affairs medical treatment facilities in Western Washington.
Telemedicine applications targeted in Project Seahawk include telementoring (teleeducation of primary care physicians, residents, medics, and nurse practitioner, teletraining, telemonitoring, and telesupport of medical equipment operation and maintenance) , teleconsultation (among physicians and other health care providers in various locations throughout the Puget Sound area and elsewhere for time-critical trauma care as well as cases requiring a second opinion, e.g., ultrasound, pathology, endoscopy, and dermatology) and telediagnosis.
Unlike other telemedicine projects, the aim of this project is to yield a system with routine clinical usefulness rather than just a technology demonstration. This demands high performance and quality as well as the most stringent design requirements. Project Seahawk also seeks to leverage existing technologies and infrastructures to minimize development cost and time.
Most existing telemedicine systems are targeted to provide teleconferencing capabilities between health care professionals and are therefore based upon commercially-available teleconferencing systems1. A general video teleconferencing system is designed to allow a group of geographically separated people to communicate with each other using primarily video and audio data. This sort of system has become common in business settings in the last five years. Often, a session is conducted with relatively poo r video/audio quality and there is no provision for displaying high-resolution, high-quality images, and archiving the session except possibly on a videotape recorder.
In reality, a telemedicine system has more unique requirements than are needed for teleconferencing. Often, high-quality transmission of video, audio, and images is required. In the case of real-time consultation of obstetrical ultrasound of potentially high-risk pregnancies, the ultrasound study consisting of images, color flow, and Doppler spectral and auditory information of good quality needs to be transmitted in real time. For dermatology applications, a high-resolution camera with either a low frame rate or still image capture capability rather than standard video at 30 frames/sec might be required. Many image processing and graphics functions are often necessary when analyzing medical images to make a primary diagnosis or plan a treatment. This ranges from window and level adjustment, magnification and minification, digital magnifying glass, image mensuration, adaptive histogram equalization, unsharp masking, and convolution to 3-dimensional visualization, texture measurements, volume measurem ents, spatial registration, lung nodule screening, microcalcification detection, and stereotactic surgical plannning.
Therefore, telemedicine systems require programmable video, audio, image handling and compression to support applications ranging from typical video teleconferencing to providing "diagnostic-quality" video, audio and medical images interactively. The acquisition, compression and processing, communication interface and transmission sections need to be tightly integrated to provide the necessary system efficiency. Support for acquiring data from electronic stethoscopes and other equipment and images from medical imaging equipment and high-resolution cameras, as well as high resolution display and processing is also critical.
The system should be able to support varying telecommunication bandwidths ranging from 64 kbps to 155 Mbps or higher depending on the clinical application, the available telecommunication channels, and the desired interactivity. For example,. teleeducat ion and basic teleconsultation applications can most often be accomplished at low bit rates with available compression and processing hardware such as the MediaStation 5000 described below. However, real-time telemammography consultation and diagnosis ap plications require higher bit rates and specialized hardware. The more challenging and difficult the remote consultant and diagnosis, the higher bandwidth and higher processing power the clinical application will require. When interactivity is either un available or unnecessary, the system should be capable of storing consultation sessions and forwarding them to the remote expert.
A telemedicine workstation prototype was developed at the University of Washington to satisfy the requirements of Project Seahawk as described above. The prototype was designed to be networked and operated in a peer-to-peer configuration as shown below in Figure 1. Each workstation consisted of a dual-monitor, 80486-based host with at least 16 MB of RAM, a 300 MB hard drive, a MediaStation 5000 multimedia card and a Fore ATM network adapter card. The host PC used Windows NT for its graphical user in terface on one monitor while the second monitor was used for image and video presentation and was controlled directly by the MediaStation 5000.
Two such workstations were developed and were installed in the Image Computing Systems Laboratory at the University of Washington and in the Department of Radiology at Madigan Army Medical Center, approximately 50 miles apart. Each system was connected via fiber-optic cable to a Fore Systems ATM switch in each location. Each switch was then connected via coaxial cable to a US West multiplexer/converter at each site which was connected via fiber optic cable to the LION Sonet ring and the other site.
Asynchronous Transfer Mode (ATM) is an emerging high-performance networking technology which combines the bandwidth guarantees of time division multiplexing with the efficiency of standard packet switching2. The use of fixed-size packets (cells) of 53 by tes enables very high-speed switches to be designed using pipelined, scalable architectures. Each switch assigns incoming cells to cell slots dynamically to efficiently use the available bandwidth. Although cells carry addressing information, called con nection identifiers, which are used to switch the cells to the appropriate circuit, ATM is inherently connection-oriented in that both the delay and the bandwidth are guaranteed for each virtual circuit. These guarantees make ATM an ideal technology for combining voice, video and data services concurrently and efficiently on a single network. And the scalability of ATM makes it useful both in LAN and WAN environments, allowing ATM to be used exclusively from end to end independent of the distance traver sed and across links operating at all bitrates from 25 Mbps to 622 Mbps and upwards.
ATM was selected as the networking standard for the telemedicine system based on the high bandwidth capabilities and the provisions for guaranteed bandwidths and maximum delays. It is assumed that the current high cost of ATM will decrease significantly in the next few years as the telecommunications industry seeks to consolidate existing voice, data and video networks into a single ATM network.
Figure 1 Project Seahawk Phase 1
Each telemedicine workstation included a Fore ESA-200PC ATM adapter card. These cards are designed for the EISA bus and, are therefore limited at the host interface by the 33 MHz bus speed. The network interface is a Transparent Asynchronous Transmitter/Receiver Interface (TAXI) with a maximum 100 Mbps transfer rate. Each ATM adapter was connected via fiber to a Fore ASX-200 ATM switch. These switches are each roughly the size of a PC and are capable of networking up to 16 devices, supporting up to 12 full OC-3 (155 Mbps) circuits simultaneously. Each switch included a 4 port LAN module and a 2 port DS-3 (45 Mbps) WAN module. The switches required minimal configuration beyond th e assignment of Internet Protocol addresses. All the ATM circuit connections were set up automatically, switched dynamically, and terminated automatically.
The Local Exchange Carriers (LECs) serving western Washington State have cooperatively developed an advanced network known as the LATA Integrated Optical Network (LION). It is an optical fiber ring designed to provide high bandwidth and reliability. Most sites included in Project Seahawk can be served easily by the LION ring. Those not reachable via the LION network will require alternative networking strategies and will likely be limited to reduced bitrates such as T1 (1.54 Mbps) or ISDN (128 kbps). Interfaces from the LION network to dedicated T1 connections or ISDN sites are available through the local carriers.
Although the connections to the LION Fiber Ring were limited to DS-3 speeds (45 Mbps) for the demonstration, the ring is capable of supporting OC-3 speeds (155 Mbps) and higher.
The MediaStation 5000 (MS5000) was chosen to provide the necessary compression and image, video and audio processing for the telemedicine workstation. The MS5000 is the latest entry in a series of high-performance medical imaging workstations developed a t the University of Washington's Image Computing Systems Laboratory (ICSL)3. The MS5000, which uses either a 486 or Pentium-based PC as a host, is a highly-integrated, programmable multimedia system in that, with the capability of 2 billion operations per second, it can handle and process medical images, live video and audio, graphics and text. The MS5000 can perform real-time MPEG-1 video/audio encoding and H.261 encoding and decoding, and fast image display and processing, e.g., a window and level ope ration on a 1024x1024x12-bit image can be completed in less than 32 ms and a 3x3 convolution on a 512x512 image takes 19 ms.
There are a number of standards to choose from when designing a telemedicine system. Each of these standards apply to one or more layers of the typical network protocol stack as shown below in Figure 2.
Figure 2 Telemedicine Protocols
The H.320 family of international teleconferencing standards provides for simultaneous audio (G.700), video (H.261) and data transfer (T.120) using communication bitrates from 56 kbps to 1.92 Mbps4. Automatic negotiation between connected sit es through H.221 and H.242 allows dynamic assignment of bits to individual audio and video channels based on the multimedia capabilities at each site and the available bandwidth. Additional connections can be established as more bits are required and both audio and video compression rates can be adjusted up and down to match limited bitrates. Compatibility with H.320 insures interoperability with the widest range of third-party teleconferencing systems. H.320 is designed to work with the range of bitra tes available using ISDN.
The MPEG-1 and MPEG-2 standards define methods for compressing high-quality audio and video. MPEG-1 supports compression of VHS quality video and CD quality audio into a 1.5 Mbps bitstream or higher quality at proportionally higher bitrates5. MPEG-2 supports compression of wide ranging quality video and audio beyond that of HDTV into a bitstream up to 100 Mbps6.
ATM is a high performance networking protocol designed to support the widest range of telecommunications data and the ATM Forum has established a family of ATM Adaptation Layers (AALs) for the various types of data that will be carried over ATM networks. AAL1 and AAL2 include end-to-end timing information with AAL1 supporting constant bit rate (CBR) traffic and AAL2 supporting variable bit rate (VBR) traffic. Because end-to-end timing adds additional overhead and is not clearly required by telemedicine applications, we have only considered the remaining adaptation layers which do not include timing information. AAL3/4 supports variable bit rate and is optimized for compressed, continuous data streams such as video or audio. AAL5 is a simplified adaptation layer designed for maximum efficiency and compatibility with other LAN protocols such as TCP/IP.
The Internet Protocol (IP) family of protocols forms the common language for the global packet-switched network known as the Internet. IP is inherently connectionless-based. Transport Control Protocol (TCP) supports error-correction, packet ordering and acknowledgments. User Datagram Protocol (UDP) provides minimal overhead but without any of the benefits of TCP listed above, though with a resulting increase in efficiency. Point-to-Point Protocol (PPP) supports the layering of IP and other packet-swit ched protocols on connection-oriented bitstreams including typical DS-0 serial lines and ISDN.
MPEG-1 was used for video compression in order to maximize video quality for diagnostic video (e.g. ultrasound) within the available processing and bandwidth limitations. Audio was sampled in stereo at 44.1 kHz using 16 bits/sample in both directions to provide CD-quality two-way audio for teleconferencing. Data, including static medical images, and system layer messaging was transferred uncompressed. Each of these data streams (video, audio, data, system layer) was assigned to an IP socket and transf erred using TCP/IP layered over AAL5.
The graphical user interface was implemented using Microsoft Windows NT 3.5 and the object-oriented Microsoft Foundation Classes of Visual C++. The menus and toolbars are available to both users simultaneously and are synchronized so that the available f unctions and images are presented identically on both computers. The toolbars provide quick access to the common image processing functions including zoom, shrink, pan, window/level, cine, synchronization of multimodal images, horizontal and vertical fl ip and 90( rotations.
As an image set is opened at either site, it is immediately transferred to the remote site and the two images sets are synchronized. Two separate cursors, white and gray, are controlled by the local and remote users respectively. Real-time image proces sing functions such as window/level and panning can then either be controlled locally or remotely. Local control provides real-time feedback to the user and updates the remote display at the completion of each operation (e.g. release of a mouse button in window/level). Multiple image sets can be open at once and real-time video can be transferred simultaneously with image manipulation..
On January 13, 1995, a telemedicine demonstration was conducted between the University of Washington and Madigan Army Medical Center using the telemedicine system prototype. Many physicians from the University of Washington, Madigan Army Medical Center a nd the Seattle Veterans Administration, were present at both ends as chest X-ray, CT and MR images and ultrasound video were exchanged, manipulated and discussed between the two sites. Response from the physicians observing the demonstration was positive with considerable anticipation for using an improved version of the system clinically. The areas that can be improved are discussed later in this section. All physicians questioned felt that the video quality was satisfactory for ultrasound consultation and that the quality of the medical images was excellent. However, audio delays averaging 0.5 seconds and up to one second were long for casual consultation and image transfer times of 0.33 seconds per 512x512x12 bit CT image and 6 seconds per 2Kx2Kx16 bit CR image could still be improved. By compressing these images, these transmission times can be reduced by a factor of 5.
Figure 3. Graphical User Interface
Image display and processing functions performance by the MediaStation 5000 was accepted as excellent. The performace of the MediaStation 5000 is described in detail in Parsons et al.7 The real-time 16-bit window/level, zooming and panning and cine operations were each executed quickly enough to maintain a high level of interactivity to the local user. The remote display, however, should be updated more frequently in the future.
In tests of network performance, maximum throughputs using TCP/IP and UDP/IP were measured at roughly 13 and 15 Mbps, respectively. These throughputs varied considerably over time apparently as a function of the availability of the host processor. Two anti cipated improvement to the current telemedicine system will be upgrades from the current VESA local bus to a PCI bus, and from the current 80486 host CPU to a Pentium. PCI is capable of higher transfer rates than VESA and will be supported in more architectures. Pentium processors offer up to 100% improvement in host processing power over 80486s, which will substantially increase the overall network performance by removing a bottleneck from the host.
Other necessary improvements for the revised telemedicine system include the use of a new Multimedia Video Processor with a higher clock speed to speed up compression and image processing, reduce delay and increase interactivity in the system. It will al so improve support for multiple simultaneous media streams. Combined with an improved video digitizer, it will yield higher-quality video. In addition, video/audio conferencing at low bitrates as well as interoperability with other teleconferencing syst ems from different manufacturers can be provided through support of the H.320 codec standard. Finally, tighter system integration will be required to increase reliability and usability.
Currently, expectations of telemedicine are unreasonably high. The smooth integration of multiple real-time media streams is still a technical challenge and the required telecommunications infrastructure is still maturing, especially in rural areas.
The high initial costs of telecommunications equipment and, in particular, the reoccurring costs of telecommunications services are currently major issues for most organizations considering telemedicine. This is further complicated by the questions of ho w to bill for telemedicine services and if they should be reimbursable. As yet, there is still no conclusive evidence to support the many claims and possibilities that telemedicine has been portrayed to deliver.
The legal implications of using telemedicine have not yet been fully explored and resolved. Physicians practicing in multiple states are required to be licensed in each state. Patient confidentiality needs to be maintained across public networks.. And the question of malpractice liability in the case of misdiagnosis and misconsultation using telemedicine is difficult.
Future research is also necessary in a variety of related issues including user interfaces, compression, medical equipment interfaces, and adaptation for specific clinical applications.
For a successful telemedicine system, various media (images, video, audio, graphics and text) need to be seamlessly integrated and supported in a single workstation. Due to its programmability and high-performance, the MediaStation 5000 is well suited fo r telemedicine and medical multimedia workstation applications. In telemedicine system implementation, one needs to take phased approach to provide a clinically successful telemedicine system with broad clinical usefulness that can grow with future needs and better technologies.
The need for telemedicine has existed for a long time. Though experiments with telemedicine date back to the 1960's, recent enabling technologies in networking such as ATM and multimedia systems with compression such as the MediaStation 5000 have reduced the costs of assembling and operating a telemedicine system significantly. However, there is still no proof that telemedicine can be cost-effective over a broad set of applications.
2 C. Patridge. Gigabit Networking. Addison-Wesley Publishing Company, Reading. MA, 1994.
The opinions or assertions contained in this article are the private views of the authors and are not to be construed as official or as reflecting the views of the Army Medical Department, the Department of the Army, or the Department of Defense.