Communication technologies in health care environments

Communication technologies in health care environments

International Journal of Medical Informatics 52 (1998) 61 – 70 Communication technologies in health care environments Andreas S. Pombortsis * Departm...

217KB Sizes 0 Downloads 22 Views

International Journal of Medical Informatics 52 (1998) 61 – 70

Communication technologies in health care environments Andreas S. Pombortsis * Department of Informatics, Uni6ersity of Thessaloniki, 54006 Thessaloniki, Greece

Abstract A successful implementation of hierarchical health care networks is dependent on a number of different technical and non technical factors. This paper investigates how the networking technologies, both ‘traditional’ and broadband, can support the communication requirements of medical environments. The general purpose is to implement communication infrastructures and services, which will improve the collaboration between the different partners in the health sector in offering better healthcare services at affordable costs. © 1998 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Medical informatics; Health care networks; Telemedicine; Broadband networks; Communication requirements

1. Background In the last few years, information and communication technologies (ICTs) have seen enormous growth and have been introduced by various degrees into the medical environment [1,2]. Each year computers are getting faster, smaller, and cheaper. The extra processing power and facilities open up the scope for much more powerful processing and networking of medical applications. Communication networks are becoming increasingly

* Tel.: +30 31 998045; fax: +30 31 998419; e-mail: [email protected]

large in size and heterogeneous in nature. New solutions knocking on the door and the new ICTs are at the center of change and influence the market transformation. Recent advantages in communication technologies have contributed to an explosion of new products and services directed at the medical environment [3]. Fig. 1 gives an abstract mapping between medical applications and network evolution. The general goal of using communication technologies in medical environments is to improve the overall quality of healthcare at an affordable cost. This requires close interaction between health care practitioners and information technologists to ensure that the

1386-5056/98/$ - see front matter © 1998 Elsevier Science Ireland Ltd. All rights reserved. PII S1386-5056(98)00125-7


A.S. Pombortsis / International Journal of Medical Informatics 52 (1998) 61–70

Fig. 1. An abstract mapping of medical applications to network evolution.

proposed technologies satisfy current user’s needs and anticipate future ones. In this direction, the paper surveys the available and viable networking alternatives which can support the networking requirements of medical applications and reports on two projects aiming to the implementation of integrated regional health care networks with a European perspective [4,5].

2. The selection framework The framework for our discussion can be defined as a three-dimensional space with the following axes: x

y z

the geographical coverage and the level in the hierarchy of given health care networks the medical applications and their networking requirements the available communication technologies

2.1. A health care network hierarchy A health care network hierarchy can be defined as follows: Level 1: Departmental networks supporting departmental information systems (e.g. laboratory, radiology, operating theater, intensive care, pharmacy, nursing, management, etc.) Level 2: Inter-hospital networking backbone(s) which interconnects a federation of distributed departmental information systems interacting with each other, supporting the integrated hospital information system (IHIS), with interfaces: (a) to communication infrastructures supporting telemedicine applications; (b) to intra-hospital HIS; (c) to PACS; or (d) to other applications. Level 3: Regional health care networks. Level 4: National health care networks. Level 5: European health care networks (networks and accompanying common activities towards a European health care network).

A.S. Pombortsis / International Journal of Medical Informatics 52 (1998) 61–70

2.2. Medical applications and their networking requirements Medical informatics has emerged as a discipline ‘that deals with biomedical information data, and knowledge and their storage, retrieval, and optimal use for problem solving and decision making’. Telemedicine can be defined as the provision of health care services, through a combination of ICTs and multimedia technologies, no matter where care providers, patients, health care records or equipment are located. Beyond the general definitions, it is well known that the classical forms of health care delivery are changing, leading to a huge number of network-based medical applications, e.g. medical imaging, patient records/multimedia medical records transmission, teleradiology, telepathology, tele-endoscopy, groupware decision support, patient monitoring, radiotherapy planning, distance learning/education, etc. [6]. The networking requirements are largely dependent on the type of application being addressed. The basic networking requirements include the bandwidth demands, the transmission delays (in accordance with the traffic type that the application generates, i.e. (1) asynchronous; (2) synchronous; and (3) isochronous), synchronization requirements (especially for multimedia medical applications), support for multipoint communication and reliability [7,8]. With regard to the cumulative communications requirements, the medical applications require support not only from the networks, but also from the protocols of higher layers (for example issues associated with communication formats, network and user interfaces, compatibility, security, etc.). Thus, the communication infrastructure should not only include networking aspects, but also software layers. Together, they determine the type of data that may be transmitted, the speed and


efficiency of communications, and the kinds of applications that an infrastructure may support.

2.3. A6ailable networking alternati6es Given the previously described networking requirements, the question of which technologies are available (and viable) and which combination of technologies is best suited to a communication infrastructure arises? It is obvious that a communication infrastructure, in order to cope with medical applications, must be flexible. This would allow the best cost/performance scenario for the various applications and environments [10]. Technology exists and it satisfies most of the needs met in the field of healthcare, however, the cost effectiveness has yet to be proven.The principal LAN/MAN/WAN networking technologies are as follows:

2.3.1. LAN options (physical layer specifications)

IEEE 802.3

10BASE5, 10BASE2, 1BASE5, 10BASE-T, 10BROAD36, 10BASE-F, 10BASE-FO, 10BASE-FA (10BASE-FB & 10BASEFL), 10BASE-FP}. IEEE 802.4 {TOKEN BUS-Broadband, Carrierband, Optical fiber. IEEE 802.5 {TOKEN RING Unshielded/Shielded Twisted Pair (UTP/STP). IEEE 802.9 Isochronous Ethernet, IVD Integrated Voice Data LAN IEEE 802.13 Fast Ethernet 100BASE-T IEEE 802.12 100BASE-VG AnyLAN, Demand Priority LAN (100 BASE Voice Grade at 100 Mbps over UTP [3–5]).


A.S. Pombortsis / International Journal of Medical Informatics 52 (1998) 61–70

Fig. 2. LAN options (physical layer specifications).

2.3.2. LAN/MAN options

IEEE 802.6 DQDB—Distributed queue dual bus ISO 9314 FDDI Fiber Distributed Data Interface FDDI-II FFOL-FDDI Follow-on LAN Fig. 2 presents the LAN/MAN options based on the physical layer specifications.

2.3.3. WAN options 1. 2. 3. 4.

Dedicated transmission lines X.25 Packet-switching Frame relay on 2 Mbps links IP packet switching networks (with Mbone the multicast extension defined for IP). 5. EURO-ISDN (with ISDN services, as it has been decided by the EURO-ISDN MoU).

6. Satellite communications (geosynchronous satellites, point-to-point fixed radio link satellites, VSAT (very small aperture terminals) supporting both ISDN and the Internet, direct broadcast satellites, mobile and personal communication satellites LEO (low-earth-orbit) satellites 7. Cellular radio technologies (advanced mobile phone system, total access communications system or TACS, group special mobile or GSM, digital cellular system or DCS 1800), 8. Radiotelephony on airplanes (the TFTS European system, The NTT-ATS Japanese system, GTE-Airfone American system). 9. BISDN/ATM (155Mbps CPNs and 34Mbps ATM cross-connect where available), wireless ATM.

2.3.4. Internetworking options “ “

Repeater (forwards bits) Bridge (forwards media-level packets)

A.S. Pombortsis / International Journal of Medical Informatics 52 (1998) 61–70


Router (forwards/routes ‘protocol-level’ datagrams) “ Brouter “ Gateway

3. Network evaluation and implementation issues Generally, the broad range of choices makes the selection of a LAN or set of LANs and/or WANs difficult. The evaluation of the existing networking technologies is based on a number of factors such as: 1. Support of bandwidth requirements (multiple access scheme, number of stations supported, fairness, bandwidth management schemes, number of multimedia streams supported). 2. Support of real-time applications (delay characteristics/provision of access-priority mechanisms/synchronization/provision of delay guarantees). 3. Provision of multicasting support (number of multicast addresses). 4. Compatibility. 5. Scalability/expandability. 6. Reliability. 7. Complexity. 8. Wide availability. 9. Cost. The most important considerations for the development of a successful communication infrastructure in a medical environment are as follows: 1. Determination of the basic needs, definition of the offered services and applications. 2. Distinction among real-time and rate-oriented applications and non-real time and unit-oriented applications. Translation of services and applications to communication requirements.


3. Definition of geographical distribution and constraints. Determination of the ‘islands’ with special networking requirements. 4. Exploitation of previous works (results from relevant project, similar installations, etc.) 5. Matching of the networking requirements against potential networking solutions/ products and compilation of cost figures for each alternative. 6. Development of checklists including all the necessary techno–economical characteristics for the: “ “ “ “ “ “ “

Departmental networks. WAN interfaces. Regional health care networks. European-wide health care networks. Inter-hospital backbone(s) networks. Intra-hospital networks. National health care networks. Usually, a HIN is supported by a campus area network (CAN), which is subdivided into subnetworks due to geographical, administrative and addressing reasons. Subnetworks are composed of segments; a segment is a shared channel interconnecting a number of users. The subnetworks are interconnected by a multi-protocol backbone, which might be, depending on the communication requirements, a switching hub, a FDDI ring, or an ATM switch-based backbone. The backbone network carries the aggregate traffic from the sub networks, while the basic interconnection rule is ‘Route if you can, bridge if you can not route’. Finally, the CAN must be equipped with various LAN/WAN interfaces for IP, Frame Relay, SMDS and ATM networks. The activities and the trials in the area of enterprise networking [11] are very important and helpful in implementing such networks.For this general scenario, the following is noted:







A.S. Pombortsis / International Journal of Medical Informatics 52 (1998) 61–70

Depending on the networking requirements, one can use intelligent hubs supporting ATM or ATM switching nodes (all in one box) leading to virtual LAN solutions. Multimedia medical workstation or hubs which generate heavy traffic can be connected directly to an ATM switch over dedicated fiber links at speeds up to 155 Mbps. It is possible to ensure that users aren’t shorted when it comes to bandwidth by introducing microsegmentation in existing LANs. Microsegmentation means assigning fewer users per LAN segment. This approach eliminates, in some cases, the need for a separate network. In some cases, the use of switched Ethernets may be less costly than to built an ATM network. The existence of ISDN based telecommunication networks may in short-terms support a sufficient number of medical applications.

4. Forecasting demand for broadband communication services The general problem of forecasting demand for broadband communication services raises the following question ‘How can we forecast the future demand that the broadband network will be required to handle, when we lack historical data (measurements) for the services we need to forecast?’ Recently, a number of forecasting methodologies and trial activities have been presented in the bibliography (RACE R2091 project, RACE II TITAN project, Bellcore Data Market Demand Model (BDMDM), Aurora, Blanca. Casa, Nectar, Vistanet, MAGIC), [13]. Fig. 3 describes a forecasting procedure.

Some basic issues associated with high speed networks, are given below [12]. “ There is a continual pressure to modernize the networks. “ The invested base must be protected “ To the end user, the technology is less important than the service it provides. Need for communication support for services that meet the evolving needs of real users in a wide range of real locales. Each of these applications may be rather simple deployment of existing technology, if they are observed in isolation. What is novel is their integration, and the impact on the user’s daily routines. “ The deployment of an integrated broadband platform, that is able to support a variety of services, minimizes capital expenditures and operational costs eliminating duplicate transmission and switching facilities, staff, operations systems, maintenance procedures, etc. “ The fate of a new technology in the telecommunication marketplace depends on the commercial availability of equipment, networks and network services which make use of this technology and on the acceptance of the technology by the users. “ Advances in existing technologies may extend the life cycle of existing network and services and slow the acceptance of new technologies.

5. The Internet The Internet is a large network (a global community) made up of a number of smaller networks whose users interact through many different applications. One of the keys to the success of the Internet has been the improvements in electronic publishing and in particular the WWW (World Wide Web). It is well

A.S. Pombortsis / International Journal of Medical Informatics 52 (1998) 61–70


Fig. 3. A procedure for forecasting broadband communication services demand.

known that the Internet is the most useful vehicle for acquisition of information (research, libraries and resources) and for interacting with others on it (e-mail, conferencing, IRC (Internet relay chat), MUD (multi-user dialog/dimension/dungeons, etc.). Also, the Internet has become one of the main supporting technologies for distance learning (lifelong learning and self-study) [14]. A lot of medical applications are based on information systems which combine IP networking with databases, WWW browsers and servers. At the moment, the general lack of wide-area bandwidth results to poor quality of audiovisual transmission over Internet. However, it is expected that the migration to high speed

networking and the next-generation protocol (IPv.6) will boost the Internet’s real-time capability as well as certain security functions. IPv.6 is designed to run well on high-performance networks (e.g. ATM) while still providing efficiently for low bandwidth links [9].

6. Network strategy based on ATM The target of B-ISDN is to merge the disparate set of networks that exist, today, into a single, unified infrastructure capable of supporting all types of communication services (i.e. support of multimedia applications


A.S. Pombortsis / International Journal of Medical Informatics 52 (1998) 61–70

in addition to voice, video, and data applications). B-ISDN is based on a cell-based transport technology called asynchronous transfer mode (ATM). Its primary goal is to provide a single underlying switching and transport technology to support a variety of rates and applications [15]. ATM is the preferred technology in medical communications. Many medical applications would require the higher bandwidth and guaranteed qualities of service supported by ATM. Interest first came from the carriers and the manufactures of WAN equipment’s, and now interest is growing in the application of ATM technology to the local (LANs) and campus area (CANs) networking environments, as well as to desktop area networks (DANs). The ITU-T B-ISDN service description lists the following classes of services: 1

Conversation services (interactive video, audio, text/data, image) 2 Messaging services (video, audio, text/data, image messaging) 3 Data retrieval services (video, audio, text/data, image retrieval) 4&5 Distribution services with and without user control (video, audio, text/ data, image distribution) 6 Computer communications (aggregate LAN, remote terminal, RPC, distributed file service, computer process swap) It is worth noting that the introduction of B-ISDN affects services, end user protocols, networked computing issues, network control and signaling, and network equipment’s and it is not simply the introduction of a new transport method. Thus, the transition to BISDN from current networks, represents a major technological discontinuity for the communications world. These discontinuities along with the competitive narrowband ser-

vices (which may reduce the immediate need for broad band services for a large part of the market) must be considered for the successful evolution of existing networks to broad band networks. In a typical scenario, there are several different networks carrying voice, data and video. An ATM based network can be used to consolidate them into one network, with the initial candidate networks being data and video, with voice networks being the last. Consolidation of various existing networks into an ATM network will depend on a number of factors such as: widespread deployment of ATM networks, development of standards for interconnection of various existing services across ATM networks, possibility of provision additional functionality and implementation of customer network management concepts described under PN and PVN services, cost savings to the customer The evolution path to a ‘pure’ ATM world could include the following phases: Phase 1:

Phase 2:

2.1 2.2

Implementation of ATMcustomer premises networks (ATM-CPNs). It can be anticipated that a CPN, according to a given requirement survey, will begin with the implementation of local workgroups and ATM-based LANs. ATM becomes the backbone. This phase has the following sub phases: Use of VP cross-connect network In ATM-CPN, signals from PBXs and LANs will be multiplexed in ATM to utilize the costeffective ATM leased-line.

A.S. Pombortsis / International Journal of Medical Informatics 52 (1998) 61–70


2.4 2.5

Introduction of broadband terminals (B-TEs) in ATM-CPN Coexistence of VC and VP networks Fully ATMizing and introduction of multimedia terminals

A gradual process, from N-ISDN to ATM private network services, then to ATM public switched services, will support the cost reduction of ATM interfaces or terminals and promote the evolution toward B-ISDN.

7. Towards to widespread use of telematic services in health care When applying communication technologies in Health Care Environments, the basic goals are to improve access to care and to enhance overall quality (by increasing the availability of some applications, improve the quality of others and facilitate some completely new ones), and to improve the medical education at affordable cost. The widespread use of telematic services in health care will grow slower than expected until several key issues, both technical and non technical, are addressed and resolved. Some principal prerequisites include: “ The cost effectiveness has to be proven. “ Acceptance of ICTs must be proceeded smoothly. “ Communication standards and formats for medical environments must be established. “ Medical applications must be re-designed from standalone to networked. “ Problems of security, integrity and availability of data must be confronted. “ Legal and ethical issues must be identified and resolved


The resolution of such problems by a continuous interaction between health care practitioners and information technologists, will reduce the existing barriers and will lead to a multi-layer strategy, for the implementation of health care communication infrastructures, with European perspective. In this direction, two projects aiming to the implementation of integrated regional health care networks are introduced. In the MIE Proceedings of the ‘RHINEAM An inter-regional health information network for Europe’ (P.H. Ketikidis et al., [4]) describes the implementation issues of a representative group of Regions called RHINE, which will be formed in order to collaborate on a trans-European basis, in the field of Information Technology applied to the health system. The long-term goal of RHINE is the implementation of an integrated distributed information system for data and knowledge interchange among users at different levels (medical doctors/patients) and in various geographical locations all over the regions. In the second paper ‘Towards healthcare digital cities. Bergamo: a best practice integrated pilot site’ (R. Alfieri et al. [5]) presents an example of integrated pilot site of several innovative projects which addresses the challenges of the fragmented health care sector by offering widespread use of telematics and simultaneously support re-engineering of the work processes. The local initiatives are integrated in the direction of building a new client-oriented cost effective healthcare enterprise with concrete benefits. References [1] M.F. Laires, M.J. Ladeira, J.P. Christensen (Eds.), Health in the new Communication Age, IOS Press, Amsterdam, 1995.


A.S. Pombortsis / International Journal of Medical Informatics 52 (1998) 61–70

[2] J. Brender, J.P. Christensen, J.R. Scherrer, P. McNair (Eds.), Medical Informatics Europe’96, Proceedings of the MIE 96 Congress (A and B), 1996. [3] IEEE Communications Magazine, special issue on ‘Communications in the Medical Environment’, vol. 34, no. 7, July 1996. [4] P.H. Ketikidis, B.V. Ambrosiadou, A. van der Werff, N. Maglaveras, C. Pappas, RHINE-AM An Inter-Regional Health Information Network for Europe, Proceedings of the MIE 97 Congress, 1997, pp. 94–98. [5] R. Alfieri, G. Brembilla, G. Orsi, L. Ricciardiello, K. Bernstein, H.B. Jensen, M. d’Angelantonio, Towards heathcare digital cities. Bergamo: a best practice integrated pilot site, Proceedings of the MIE 97 Congress, 1997. [6] A. Hasman et al., Education and Training in Health Informatics in Europe: State of the ArtGuidelines-Applications, IOS Press, Amsterdam, 1995. [7] S. Akselsen, A.K. Eidvik, T. Folkow, Telemedicine and ISDN, IEEE Commun. Mag. 31 (1) (1993) 46–51. [8] H.J. Stu¨ttgen, Network evolution and multimedia



[10] [11]



[14] [15]

communication, IEEE Multimed. 2 (3) (1995) 42– 59. T.E. Bell, J.A. Adam, S.J. Lowe, Communications-Technology 1996 Analysis and Forecast Issue, IEEE Spectrum (1996) 30 – 41. European Information Technology, Observatory 95, EITO, 1995. IEEE Communications Magazine, special issue on ‘Enterprise Networking’, vol. 34, no. 1, January 1996. A.S. Pombortsis, Some thoughts about the techno – economic comparison of communication technologies and architectures. From theory to practice and vice versa, Proceedings of Architecture and Protocols for High-Speed Networks, Dagstuhl Seminar, 1995. IEEE Communications Magazine, special issue on ‘Traffic and Service in the Years 2000 – 2005’, vol. 33, no. 2, February 1995. H.J. Ellsworth, Education on the Internet, SAMS Publishing, 1994. D. Minoli, Enterprise Networking: Fractional T1 to SONET, Frame Relay to BISDN, Artech House, 1993.