Ebook: Integration of Health Telematics into Medical Practice

In September 2002, telemedicine experts from all over the world gathered in Regensburg, Germany to join the 7^th International Conference on the Medical Aspects of Telemedicine with the topic Implementation of Health Telematics into Medical Practice. This conference was hosted by the Department of Trauma Surgery of the University of Regensburg, the German Society for Health Telematics and the International Society for Telemedicine.

During this conference, a unique exchange of research results took place, involving participants from 46 countries and 5 continents. AU major topics of current telemedicine applications and research activities were addressed and an exchange of experiences in almost all fields of telemedicine took place during this 4 day conference. This so-called Regensburg telemedicine conference was probably one of the most important events of the international telemedicine scene. In order to preserve some of the many very interesting topics, we are publishing this book, which addresses very different subjects from research to practice and from medicine to information technology and médical informatics issues.

In most of the European countries, the United States, Canada, Japan and Australia, telemedicine is already helping to keep the health care Systems under control, in terms of cost savings and quality assurance. This is reflected by significant funding from the govemments of many of these countries, which are more than ever under huge pressure to keep the health care system alive. However, telemedicine has not yet reached mainstream medicine. There is only limited interest among scientists to address the issue of the impact of telemedicine on the current practice of medicine. Proper outcomes research on the effectiveness of telemedical applications would strongly establish the Implementation of telemedicine in current medical practice. This is what makes this publication so valid.

Our goal is to keep up the good work, to make telemedicine more attractive than ever, use it as much as possible to improve delivery of health care and help us focus on the ethical basis of medicine: to serve the patient – as an individual and as part of our society – and that is the human aspect of modern technology.

The Editors

To establish a guideline for the use of teleconsultation, which is one of the most important applications in telemedicine beeing perfomed very frequently in daily medical practice in almost any field of medicine. Evidence: The recommendations are based on expert knowledge, because of the lack of evidence based data in the telemedical scientific literature. In addition, scientific articles of the highest level of evidence available published between 1970 and 1999 were reviewed.

Developement and consensus process: A guideline draft was prepared using the attributes of clinical practice guidelines developed by the Institute of Medicine (IOM) of the National Academy of Sciences. This draft was reviewed by telemedicine experts and the content of the guideline was approved by 100% group consensus in 5 meetings held between 1997 and 2000 by a panel of members of the Subproject 4 Group of the Global Health Care Application Project of the G8 countries.

Conclusions: The guideline gives recommendations on all aspects of teleconsultation in any field of medicine and will be updated regularly implementing new evidence.

Introduction: Every year many disasters cause thousands of injuries, deaths, refugees. Depending on the kind of disaster (train/plane accident, flood, earthquake) not only an acute emergency medicine treatment but also generai and family medicine and hospital treatment have to be safeguarded over a longer time-period in the disaster area.

Problem: Regarding to a lot of organizations, institutions and disaster teams taking part in the disaster assistance is there any lack of work or data flow in the medical treatment?

Methods: From the ODRA flood 1997, the high speed train crash in ESCHEDE 1998, the DANUBE flood 1999 and the ELBE flood in 2002 experience reports were collected. They were analysed with emphasis on data and work flow in the medical treatment and its command system: Standardised command structure? Communication problems? Used communication lines? Language problems? Médical Intelligence distribution? Use of Patient Tracking System? Triage problems?

Results: The use of spoken radio communication causes transmission mistakes or misunderstandings and radio-overload and need connection-set-up-time for each call. Manual distribution of same data for many receivers using different communication lines causes a time shift in the up-to-date-information. Language problems during the ODRA flood between German and Pohsh people led to longer reaction times. Up-to-date triage results as weil as up-to-date transportation and hospital information are necessary for medical evacuation. Compared with other reports about these disasters the quality of disaster management depends on the quality of communication and information.

Conclusion: The use of health telematics in disaster response helps to cope with the scenario. Modern technologies provide support for building up medical aid although the normal infrastructure is destroyed. To cope with disaster scenarios there are some telematic tools which can be used:

- Computer-based Command and Control System

- Telemedical support

- Data-ressources-network / Medical Intelligence

A further study is recommended to evaluate the real impact of using these telematic tools in a disaster.

Electronic Health Record (EHR) Systems provide the kernel application of health information Systems and health networks which should be independent of complexity, localisation constraints, platforms, protocols, etc. Based on shared care information Systems' requirements for high level interoperability, a generic component architecture has been introduced. For implementing, running and maintaining acceptable and useable health information Systems components, all views of the ISO Reference Model – Open Distributed Processing have to be considered. Following the Model Driven Architecture paradigm, a reference model as well as concept-representing domain models both independent of platforms must be specified, which are combined and harmonised as well as automatically transferred into platform-specific models using appropriate tools.

The integration of telecommunications and information Systems into health care delivery in human space flight operations is not new. It has been an integral tool for over 45 years. During these past decades, numerous efforts have been conducted to further develop and promulgate telemedicine. The National Aeronautics and Space Administration (NASA) established a commercial space center in 1997, known as the Medical Informatics and Technology Applications Consortium (MITAC). MITAC has developed and conducted a variety of test beds in several international settings, including Russia, Ecuador and other extreme and remote environments. These test beds have been designed to evaluate and validate technologies and techniques that have application in the delivery and support of health care in unique environments. The characteristics of these test beds are analogous to what might be observed or experienced in low earth orbit or on space-based platform. These include intermittent Communications, low bandwidth, level of competency of the front line health worker, etc. These test beds have led to new approaches for the delivery of health care as well as enhanced education. These experiences have been beneficiai in the promulgation of telemedicine as an effective tool and have provided new ideals for space exploration as well terrestrial medicine. This paper will highlight MITAC’s test beds and their relationship to space exploration.

The use of robotics in surgery is nothing new. However, there are areas of surgery, such as in fracture fixation, where robots have yet to be implemented. This paper considers the choice of robot, gripper and ancillary equipment together with navigation systems necessary for their application.

Hitherto robots have seen operation in surgery only in cases where relatively low manipulation forces are required. Nothing yet exists with the capability of handling forces in excess of 200 Newton as would be required in the above scenario. Another encumbrance to robots which are already in medical use is the difficulty in programming. Unfortunately most of these robots are programmed by specialists for a particular application. However, there exists a number of robot programming languages, like Unimation VA-LII (recently superceded by Stäubli V+), which do not require specialist knowledge. The application of industrial robots to the “heavier” side of modern surgery is without doubt technically realisable. The remainder of this research project aims to determine exactly which robots and what ancilliary equipment are needed and then to implement them, first on plastic models and later on cadavers. A second phase is expected to deal with type approval and a final third phase with operations on live patients.

Telehealth “readiness” can be defined as the degree to which users, health care organizations, and the health system itself are prepared to participate and succeed in its application. This project developed a readiness model for rural/remote locations in Canada. Specifîcally defined groups or communities with shared characteristics within a rural geographical Community (i.e. practitioners, patients, the public, and health care organizations) participated in key informant interviews, awareness sessions, focus groups, and face-to-face interviews. The data were examined and organized keeping in mind Weiss' Program's Theory of Change. This approach allowed concrete and abstract factors to be considered. The model that emerged suggests that there are four types of readiness for each of the defined communities: core, engagement, structural, and non-readiness. The “communities” share some readiness factors and risks, but also exhibit unique clements. This finding is critical to acknowledge when the goal is to implement a useful, effective, and sustainable telehealth system within remote settings. Study results hold a key to understanding why technology Systems have failed in the past, in spite of dedicating considerable human and financial resources towards their implementation. Notations of these findings will be helpful in future telehealth implementations within rural and isolated areas.

A computer-based automated histopathology recognition system was developed to distinguish benign from malignant lesions. Tubular carcinoma of the breast, which has several reactive and neoplastic mimics, was selected as a model. Archivai stained tumour sections from the United Kingdom National External Quality Assurance Scheme for breast pathology and supplementary material from external pathologists formed the study population. A diagnostic process similar to that employed by the histopathologist was adopted, viz, low-power feature extraction and analysis by cluster/glandular groupings followed by high-power confirmation. To circumvent problems of stain variability, greyscale quantisation of images was achieved through Karhunen-Loeve transformation with results suggesting that histological stains provide information primarily through contrast and not colour. Mean nearest neighbour and variance of cell nuclei distances were found to be 100% effective in distinguishing images which contained diffuse tumour, and no clustering. Gaussian smoothing followed by minimum variance quantisation allowed segmentation of gland clusters. Perona-Malik nonlinear diffusion filter employed prior to intensity thresholding and morphological filtering was 92% (7330/7973) effective in segmenting individual glands. In a set of 62 benign and 52 malignant gland clusters, the features found to discriminate tubular carcinoma from benign conditions included >20% of glands with sharp-angled edge, cluster area >150,000 pixels, ratio total gland area:total cluster area <0.14, >60 glands per cluster and the ratio average malignant gland area:benign gland area <0.5. Suspicions clusters were subjected to high-power feature analysis for nuclear morphology, nucleoli detection and basement membrane assessment. Watershed thresholding achieved nuclear segmentation and nuclear area >1.3x mean benign nuclear area was found to have a malignant likelihood ratio of 14.5. Progressive thresholding was used to detect nucleoli. Basement membrane was accentuated by colour segmentation and demonstrated 0.96 sensitivity, 0.89 specificity and 0.92 positive predictive value for distinguishing malignancy.

Modern data analysis is one of the many prerequisites for telemedical applications. Classical Statistical methods alone are no longer sufficient to fulfill the various demands of modern analytical procedures. Cluster and association analysis among others have filled this gap and are capable of producing more adequate and better suitable results as well as to provide information not detectable in the past.

Intelligent and, thus, autonomously reacting software programs are capable of handling a lot of different tasks as has been realized in economics and network administration. The same so-called software agents can be used for a variety of organizational tasks in medicine. Some software agents already manage an individual patient's health care record from documentation to ambulant or stationary admission, surgical planning, and many other tasks which currently consume more than half of a physician's daily working time. Hence, not only a large potential of time, but also of economical savings result to the physicians' new disposition.

The ability to image the elastic properties of tissue is potentially useful in a variety of applications. The field of elastic imaging has grown in response to the Potential use of such information in medical diagnosis. Real time ultrasound elastography represents a recent development in determining strain and elasticity distributions. Nevertheless, commonly used imaging techniques rely on the interpretation of two dimensional visual data displayed on a video screen. In reality however, physicians often prefer tactile exploration making the simultaneous portrayal of both video and haptic information most desirable.

Since the 1970's many alphanumeric to tactile data conversion methods have been investigated, mainly with the ultimate aim of assisting the blind. More recently, interest has been directed toward the display of pictures on haptically explorable surfaces - Tactile imaging. Such a system would allow surgeons to examine hard sectors contained within soft tissue, and thereby assist in operations held remotely. The expansion of ultrasound elastography to 3D formats would mean the ability to haptically explore regions of the body normally inaccessible to human hands. For three-dimensional imaging the acquisition of sequential tomographic slices using Elastography, combined with image segmentation, enables the reconstruction, quantification and visualisation of tumour volumes.

In a collaborative project between four research Institutes, the aim is to produce a prototype three dimensional tactile displays comprising electrically switchable micromachined cells, whose mechanical moduli are governed by phase changes experienced by electrorheological and/or magnetorheological fluids. This will be integrated with a sensory ultrasonic elastography in order to present the human fingers with controllable surfaces capable of emulating biological tissue, muscle and bone.

The National Disaster Medical System (NDMS) was created in the early 1980's, and it was designed to meet the threats of the time. Today the threats are much less discreet and predictable. They are distributed; they move and spread quickly; and they walk silently among us. Specifically, biological agents are an enemy unlike any we have had to deal with before. They offer unique challenges that fly in the face of current doctrine. We must redesign the NDMS in order to contain and eliminate this new threat. Tools exist today capable of effectively coordinating distributed resources - even through containment borders. We need to strengthen our public health system, create a net-centric disaster management system, and blur the boundaries between local and federal resources. Ultimately we must move from an incremental, echelon-based response to an immediate, continuons response. This can be accomplished by adding inexpensive, well-estabhshed information technologies to the existing response system.

Communication theory based analysis sheds new light on the use of health telematics. This analysis of structures in electronic medical communication shows communicative structures with special features. Current and evolving telemedical applications are analyzed. The methodology of communicational theory (focusing on linguistic pragmatics) is used to compare it with its conventional counterpart.

The semiotic model, the roles of partners, tiie respective message and their relation are discussed. Channels, sender, addressee, and other structural roles are analyzed for different types of electronic medical communication. The commimicative processes are shown as mutual, rational action towards a common goal. The types of communication/texts are analyzed in general. Furthermore the basic communicative structures of medical education via intemet are presented with their special features. The analysis shows that electronic medical communication has special features compared to everyday communication:

A third participant role often is involved: the patient.

Messages often are addressed to an unspecified partner or to an unspecified partner within a group. Addressing in this case is (at least partially) role-based.

Communication and message often directiy (rather than indirectly) influence actions of the participants.

Communication often is heavily regulated including legal implications like liability, and more.

The conclusion from the analysis is that the development of telemedical applications so far did not sufficiently take communicative structures into consideration. Based on these results recommendations for future developments of telemedical applications/services are given.

All industrial societies are ageing. This has profound socio-economic and health sector implications. Innovative services based on Information Society Technologies (IST), like telehomecare are regarded as promising avenues to follow both to allow (national) health Systems to cope with these challenges and to improve the quality of life of chronically ill and frail older Citizens. The aim of the TEN-HMS project is to convincingly prove that telemonitoring of congestive heart failure (CHF) patients at home can improve medical outcome for these patients as well as their quality of life and the efficiency of healthcare delivery processes. But this will not (yet) be enough for the sustained success of such a service. Unless it takes into account the interests of the various players in the health care arena and a long-term Business Case can be proven, it will be very difficult to integrate such services into routine health care delivery processes. Before developing concrete delivery models for such a telemonitoring service, the “players” directly involved in such a service need to be identified - customers/patients, health services providers, IT services suppliers, and public/private insurance funds as payers - and their assessment perspectives considered. Then four concrete telemonitoring delivery models and their probability of success are discussed. Our analysis suggests that telemonitoring will presently only be successful if the service delivery model applied reflects national health system idiosyncrasies, takes into account established organisational boundaries and adapts to patient quality of life and health professional preferences. In the longer term, the new paradigm of seamless, patient-centred care will, however, require new, more efficient service delivery models integrating all aspects of the health services value chain.

Early diagnosis is imperative in the quest for cures for or prevention of diseases. Medical doctors nowadays can measure a multiplicity of physical determinants such as blood pressure, insulin levels, etc. The sooner this is done the sooner therapy can begin and the sooner therapy begins the greater the chances of success, no matter what the disease.

That is why self-sufficient home monitoring and emergency aid Systems are becoming ever more important in medical health care.

The German emergency care system is a very sophisticated one. However, negative headlines like “emergency Patient Tourism” appearing from time to time, have provoked a thorough deficit analysis which revealed two weak points: communication and documentation. The communication system presentiy used is a rather outdated one employing analogue voice radio between the ambulance cars/helicopters and the dispatch center and telephone communication between the dispatch center and the emergency rooms. To document the emergency case the on-scene physician is required to fill out a form. A survey showed that many of these forms are filled out incompletely and/or inconsistently or are even missing completely.

NOAH, which means “Emergency Organization and Administration Aid” (“Notfall Organisations- und Arbeits-Hilfe” in German) intends to address both of these communication and documentation deficits. The on-scene physician is equipped with a mobile ruggedised PC with an internal digital radio-modem. It provides a direct, digital communication Channel from the on-scene physician to the emergency physician starting from the first minutes of the treatment of the emergency patient. It also provides an easy-to-fill-out variant of the paper form mentioned above with an on-line help function and visual aids.

A typical course of events with the communication part of NOAH is as follows. The on-scene physician is alarmed via NOAH and can obtain details of the emergency during the approach. He can enter the status codes (“on the move”,“arrived at the emergency scene”,“arrived at the patient”, etc.) with NOAH. During the first minutes of the treatment of the patient the physician enters a so called “First Message”, which requires only 10 to 15 seconds. This message contains basic information like sex, age and the injuries of the patient. It helps the dispatch center (if the on-scene physician desires so) to make an informed recommendation where to bring the patient. This message is also forwarded to the emergency room of the destination hospital, where it can help to start appropriate preparations for the patient. When the on-scene physician has enough time, he can already bring up the documentation masks and give further information to the emergency room.