Contact Person: Kevin M.
McNeill, Ph.D. (firstname.lastname@example.org)
Materials & Methods
For any telemedicine program it is important to understand the specific goals of the program to determine the technology necessary to support those goals, as well as the business aspects. Our approach was to divide technology issues into two categories: applications technology and telecommunications technology. In most cases these two technologies can be evaluated independently and involve different areas of expertise. For the ATP the selection of applications technology primarily involves medical staff and personnel from Biomedical Communications. Applications equipment raises issues of efficacy and diagnostic accuracy and is the key focus of the program's technology assesment component. We view the telecommunications technology as infrastructure which should be transparent to the end-users of the applications equipment. In the ATP telecommunications issues primarily involve network engineering personnel.
In evaluating video equipment we compared the products of leading vendors who offered telemedicine specific equipment. In addition to standard video conferencing capabilities these units are designed for medical application and provide for peripheral devices to support various types of examination (e.g. ENT scopes). These units are also designed to meet code requirements for operation in the hospital environment. This type of evaluation must include both video experts and medical personnel and should include a variety of medical specialists depending on the range of services to be offered by the telemedicine program. Once challenging application is support for the real-time transmission of ultrasound video for cardiology and natology applications. Testing of the quality of video for this application was important for our program. In addition, we found many aspects of all the products we evaluated which were unsatisfactory. An important criteria for us was the willingness of the vendor to consider changes to their equipment to meet our requirements. Our utlimate selection of a vendor for real-time video equipment was based largely on the quality of their video codec and their willingess to change their equipment to our specifications.
For store and forward equipment we evaluated leading vendors and selected PC based equipment which supports general purpose telemedicine. This equipment includes software which allows a physician to capture a patient encounter using video clips, audio clips, scanned documents or digitized film. This information is then organized as a multimedia record and sent using a software tool very similar to an e-mail application. At the receiving end an idication is given that a new case has arrived and the site coordinator can review the case and arrange for the appropriate consultation. We found however that the general purpose store and forward equipment was not satisfactory for use in teleradiology. Teleradiology has a demanding set of requirements regarding image quality [4, 5]. This was due to the quality of the image viewer and some problems with adequate control of the film digitizer. For site at which teleradiology is a primary interest we utilize a second store and foward system specific to teleradiology. These system can acquire images through video capture for CT/MRI as well as use a film digitizer. They send images to a dedicated teleradiology server. In addition, at one client site an interface was added to the CT which allows the digital data to be sent directly to the server, at full spatial and contrast resolution, using DICOM standard protocols. Viewing of the radiology images is done on a dedicated teleradiology workstation in the diagnostic department. Images are sent from the server to the view using DICOM transfers. Ultimately the teleradiology server will be able to send the images to a full scale archive and images will be accessible at any digital radiology workstation in the diagnostic department.
For the telecommunications infrastructure we evaluated Asynchronous Transfer Mode (ATM) technology and found it to be a good fit with our requirements. Recent trends in the industry had also made it much more affordable that it had been in the past. Often, ATM is associated only with expensive very high-speed communications lines such as T3 (45 Mbps) or OC3 SONET (155 Mbps). W e were able to find inexpensive end equipment from Fore Systems which supports ATM over T1 circuits and can provide dynamic allocation of bandwidth between video and data applications. In addition, the device provides a LAN interface that allows us to sup p ort data applications without requiring an external router. Therefore, instead of three or four components at the remote site (a CSU/DSU, mux, a router and possible a hub) we could provide the telecommunications infrastructure with one or two devices (an A TM T1 WAN mux and possibly a hub). This was attractive from the stand point of simplified installation and maintenance. The same vendor also offered very scalable and affordable ATM switches for the backbone which could support a variety of communications media ranging from T1 to OC12 (622 Mbps) with switching capabilities that can be expanded incrementally from 2.5 Gbps to 10 Gbps.
In the state of Arizona most sophisticated new telecommunications services are introduced in the Phoenix metropolitan area and migrate later to the smaller cities of Tucson and Flagstaff. The rural areas of the state consistently lag in the availability of all but the most basic services. Arizona is a very mountainous state and many of the rural sites are located in regions where cellular services are limited as well. In addition, cellular services will not support the bandwidth requirements of key telemedicine applications. We also investigated satellite connections but found them to be more expensive than dedicated T1 circuits for the equivalent dedicated bandwidth.
A key advantage to the ATM network architecture we have deployed is that it allows us to reduce the costs associated with the T1 circuits linking each remote client site since we can route that circuit to the nearest ATP network POP (Point-of-Presence). This reduces distances based charges. Once at the POP the digital data can ride on the existing backbone of the network to get to the hub site. This architecture ensures that the backbone is fully utilized and reduces the cost of connections to the rural sites. It also allows us to connect most rural sites without the need to cross LATA boundaries. Crossing LATA boundaries would limit the program to using long distance carriers to obtain communications services since local exchange carriers cannot haul traffic across those boundaries. By providing POPs in three regions of the state, competition among providers for the individual rural T1's is enabled, allowing us to select the lowest possible price. The backbone lines do cross LATA boundaries and are subleased by the ATP from Northern Arizona University (NAU). NAU has a backbone of T3 linking the state from north to south and is able to sublease T1 circuits to link our switches. For many of the rural sites the ATP has recently converted from month-to-month pricing to a reduced 5-year contract rate.
Back to the top.
© 1998 Arizona Board of Regents