TeleHealthcare: The interface between education, medicine and telecommunications

TeleHealthcare lies at the interface of education, healthcare and telecommunications. In many ways its arrival has more to do with advances in telecommunications than advances in computing. "TeleHealthcare" promises to radically alter the way in which health care is delivered, taught and assessed. The aim of TeleHealthcare is to improve communications between clinician, educator and patient, to increase the level of interaction and customer support, to decrease costs and response times. This can occur in all phases of medical care. TeleHealthcare particularly comes into its own in areas where geographical isolation makes face to face meetings difficult, expensive or culturally inappropriate. In Western Australia all these factors apply to rural healthcare and particularly to the provision of care and education to remote aboriginal communities. In the Department of Anatomy and Human Biology at UWA we are investigating the role of video conferencing and groupware in the education of medical, dental and science students.

The nineties have seen the emphasis in Information Technology move from hardware to communications. We are deluged with information on the Internet from news and entertainment media. I would like to concentrate on a lesser known aspect of this communication revolution, the emerging field of telemedicine. "TeleHealthcare" promises to radically alter the way in which health care is delivered, taught and assessed. TeleHealthcare lies at the interface of education, healthcare and telecommunications. In many ways its arrival has more to do with advances in telecommunications than advances in computing.

The aim of TeleHealthcare is to improve communications between clinician, educator and patient. The aim is to increase the level of interaction and customer support, to decrease costs and response times. This can occur in all phases of medical care.

Telehealthcare	Server	Client
Telemedicine	Doctor	Patient
Telepathology	Pathologist	Doctor/patient
Telecardiology	Cardiologist	Doctor/patient
Telediagnosis	Expert/Expert System	Doctor/patient/paramedics
Telepsychiatry	Psychiatrist	Doctor/patient
Telementoring	Expert	Practitioner
Emergency Services	Expert	Ambulance/paramedics
Distance Learning	Expert	Student
Nursing	Nursing Station	Nursing Station
Tele-Procurement	Wards/requester	Supplier
Remote Security	Ward	Security Staff

Telehealthcare particularly comes into its own in areas where geographical isolation makes face to face meetings difficult, expensive or culturally inappropriate. In Western Australia all these factors apply to rural healthcare and particularly to the provision of care and education to remote aboriginal communities. The average distance travelled by an Aboriginal patient to the nearest healthcare provider is 300 km. Healthcare in these situations may involve the patient in considerable expense and disruption as relatives may have to leave work, pay travel costs and find accommodation miles from home. It may also present social and ethical problems by removing the individual from their usual social and community support to the alien environment of a big city.

In many of the developing countries of the Pacific Rim, health care expertise is concentrated in capital cites. In some cases specialist knowledge may even have to be obtained from outside the country. Sharing the same time zone as many of these countries, Perth is well placed to provide this service. In Thailand and China "health messengers" go out to the villages in country areas, even local village drugstores can act as centres of diagnosis and basic healthcare if able to draw on expertise from regional centres of expertise when required.

The level of communications will vary with the size of centres involved. The centre of population (eg. a capital city) may have a large Unix based workstation, which will pass information on to regional centres running X-windows based workstations (for example). Health centres may have top end PCs running Windows NT while individual users such as GPs or the general public may use Win95 and Java. It is expected that the costs of these services and the bandwidth of traffic between them will decrease as the information passes out from the centre to the periphery.

Major centres may be linked by satellite or dedicated fibre optic cable, smaller centres may be reached by ISDN lines (effectively high quality, high speed phone lines dedicated to digital data exchange) while the small end user may rely on conventional copper phone lines. There are three main uses of such interconnectivity each requiring greater bandwidth for increasing quantities of data transfer: Voice and pure textual transfer for transfer of textual patient data; transfer of high resolution images such as X-rays and MRI scans; and live video for video conferencing , ultrasound and remote patient observation.

The main problems encountered when trying to introduce some form of telemedicine are not technical but problems with the reliability and cost of the telecommunications. The speed and cost of communication are not linearly related.

Transmit CT scan	Modem 24 kbs	ISDN 64 kbs	FastPac 2 Mbs	Fastpac 10 Mbs
Cost $A	55	137	3	5
Time taken	1.8 hrs	33 min	1.1 min	12 secs

Some of the anomalies are due to the pricing structure of telecommunications still being based on the voice era. Most people speak at roughly the same rate (approximately 100 words per minute, a transfer rate as slow as 80 bps) so that it makes sense to price by time of use. This provides little incentive for a telecommunications company to provide faster data links and has led to the anomalous situation where megabytes of Internet information can be obtained for cents over a line which may also be carrying conversations being charged at dollars a minute.

With digital data transmission speeds varying enormously depending on the hardware and protocols involved it makes more sense to charge by the amount and quality of information transferred. This would encourage the development of fast transfer mediums. As important as speed is the security and quality of the transmission. Radiographic images sent for diagnosis can only tolerate very small scale errors of transmission. On a "noisy" line this may necessitate many transmissions or much wasted overhead in checking the integrity of the transmitted image. Thus a fast but imperfect line may be less effective than a slower but cleaner line.

One of the pioneering areas of TeleHealthcare is in Teleradiology. Increasingly medical images are produced in digital format, in many cases these are rather absurdly printed on radiographic film so that it can mimic the traditional thin film X-ray. Trial links are already in place between Bunbury, Busselton and Perth. An X-ray or ultrasound taken remotely can be observed by an expert in Perth and a decision made as to whether the patient needs to travel to Perth for advanced treatment.

In this immature field there is as yet little in the way of quality control and monitoring. Issues of liability and security have yet to be addressed. There is no licensing authority and little in the way of training. There are as yet no enforced standards for image transmission but good practice suggests that large matrix images such as an X-ray require 2k x 2k x 12 bits for transmission and 2k x 2k x 8 bits for display. Small matrix images such as CT scans, MRI, or ultrasound require 0.5k x 0.5k x 8 bits minimum.

Another route for increasing communication is video conferencing. This can place even more strain on bandwidth than teleradiology. Resolution is less important but frame rates of 20 ps or greater are required to produce smooth movement. The need for duplex (bidirectional) data transfer effectively doubles the data transfer rate required. Although restrictions on available bandwidth may be temporary and we may look back on this as we do to days when all programs had to be written to run in 64 kb of RAM. we are stuck with it for the foreseeable future. Thus much effort has gone in utilising the computer for compression and decompression at each end of the bandwidth limited transmission medium.

Three main standard for moving images are JPEG, H320 and MPEG. JPEG is most suitable for LANs with asynchronous protocols as each frame bears its own identifier and the loss of frames is not catastrophic. It is also good where there is a lot of movement as full frame are sent. This contrasts with H320 which sends much less information for the same length of time as only changes between frames are transmitted. This is particularly good for images like "talking heads" where most of the image is static from frame to frame. It does, however, require accurate synchronous transmission as each image relies on the previous one for information. MPEG provides better quality sound and vision than either JPEG or H320 but requires decompression hardware in the target machine.

The image quality depends directly on communication bandwidth and infrastructure available. At the bottom end conventional telephone lines can be used either for plain voice transmission or to access the internet via a modem and gateway. In theory up to 6 Mb in and 386 kb out is possible along conventional phone lines but this relies on everything being in perfect condition. At this level video conferencing is possible but can only provide about 10 frames per second (fps) in monochrome. Such a system operating over the Internet is offered by Creative technologies and is being tested in the Department of Anatomy and Human biology at UWA. In this system the image is jerky and there may be considerable delay between speech and vision. Full colour still images can be transmitted in seconds.

The next step up is to use leased ISDN lines, linking two together can give an effective sustained transfer rate of 256 kbs. This can provide full colour duplex transmission at up to 20 fps. To provide better quality than this requires dedicated lines with great bandwidth such as a dedicated optic fibre network. Conventional Ethernet local networks can rapidly become saturated when video conferencing occurs even with up to 60% of the 100 Mbs capability available. Switched Ethernet is best for the transfer of large image files as the bandwidth does not have to be shared with all the other traffic going on at the time.

As deregulation of the telecommunications industry proceeds the cost of Telehealthcare will continue to drop. Perth is well placed to provide the training and expertise required in the region. In our department we are using an Innovations in Teaching grant to introduce video conferencing to our undergraduate medical and dental students. Computer mounted video cameras transmit monochrome images over the LAN to educators in another building across campus. This system is being installed this year and preliminary results will be presented at the conference.

Author: Stuart Bunt, Department of Anatomy and Human Biology The University of Western Australia, Nedlands WA 6907, Australia Tel 09 380 2983 Email: smbunt@anhb.uwa.edu.au Please cite as: Bunt, S. (1996). TeleHealthcare: The interface between education, medicine and telecommunications. Different Approaches: Theory and Practice in Higher Education. Proceedings HERDSA Conference 1996. Perth, Western Australia, 8-12 July. http://www.herdsa.org.au/confs/1996/bunt.html