Lade Inhalt...

The digitization of radiology and its impact on patients, processes and economics of modeling

Implementation of holistic and patient orientated information systems in teleradiology networks

Fachbuch 2013 194 Seiten

Gesundheit - Public Health



List of Abbreviations

1.2 Aims and Objectives
1.3 Work Procedure

2 State of the Art
2.1 Medical Documentation and Classification
2.2 Communications standards in Radiology
2.3 Medical Imaging
2.4 Image Compression
2.5 Picture Archiving and Communication System (PACS)
2.6 Hospital Information system (HIS)
2.7 Electronic Patient Files
2.8 Radiology information systems (RIS)
2.9 Technical Network Basis
2.10 Teleradiology
2.11 System integration architectures and models
2.12 Security and Data Protection Aspects

3 Materials and Methods
3.1 Literature Research
3.2 Genericness
3.3 Holisticity
3.4 Managing Complexity
3.5 Patient Orientation
3.6 Theoretical Modeling

4 Theoretical Analysis
4.1 The Effect of Digitization on the Patient
4.2 Effect of Digitization on Processes
4.3 The Effects of Digitalization on the Economy

5 Modeling the Implementation Method
5.1 Differentiation between Development, Implementation, Operation and Integration of Information Systems
5.2 Description of the Implementation Method
5.3 Procedure Model of the Implementation Method
5.4 Information Container as the Core of the Implementation Method
5.5 Project generation
5.6 Process Team and Project Team

6 Discussion
6.1 Meeting Objectives
6.2 Methodology
6.3 The Limits of the Concept
6.4 Comparison with other Concepts
6.5 Usability of the Concept for Adjacent Areas
6.6 Final Conclusions
6.7 Outlook

7 Summary


9 Tabular Attachment
9.1 Table of Illustrations

List of Abbreviations

illustration not visible in this excerpt



Medical technological progress, marked both by the many developments in radiology in recent decades, as well as information technology, including communication technology in recent years, has many applications and great potential for synergies, which, however, can only be realized through the comprehensive integration of the various strands. In this work the relevant factors are analyzed theoretically. From this has come the development of a method of holistic and patient-oriented implementation of information systems. The focus is on the use of information systems in teleradiology networks, the processes and all these acting factors as a basis of information technology for healthcare professionals.

The introduction of X-rays in medical diagnostics was a quantum leap for medicine with applications which were initially difficult to imagine. X-ray diagnostic procedures have been in continuous development since the discovery of X-rays more than 100 years ago. Today, the department of diagnostic radiology includes the production, processing and management of relevant medical images of patients. For most types of investigation, the radiation exposure of patients has been significantly reduced. In recent decades, the development of sensitive film-screen systems, modern image intensifier technology and digital systems have contributed to reduce levels of radiation exposure. The combination of radiology with the application of information technology has led to the digitization of radiology. The images are created digitally, stored and communicated over long distances quickly and consulted on by medical specialists. The images can be saved in structured archives and opened quickly when needed.

The digitization of radiology is associated with other information technology developments in medical applications. Picture archiving and communication systems (PACS), radiology information systems (RIS), laboratory information systems (LIS) or hospital information system (HIS) are just a small sample of the information systems used.

Because of increasing amounts of data and requirements for long-term and secure storage of patient data, such as image archiving, document management and data management in radiology and hospital information systems, ever greater demands are being made on the design of information systems in the medical field. Most medical tests consult radiological diagnoses, and, particularly in radiology, a high volume of data is recorded, described, processed and archived. In computer tomography and magnetic resonance imaging the images are already created in digital form, but classic analog methods, such as X-ray and endoscopy are also increasingly being replaced by digital recording processes.

This area of application of PACS and RIS is the growth market for information technology in health care, because it has always been the case in radiology that without the support of information technology it is often considerably more work to handle current image data, to archive it, and, if needed again, to locate images in the archives. This is not just limited to large hospital corporations, or hospitals, but is also of interest to practicing radiologists and other health professionals. These considerable time, labor, and space requirements and costs can be reduced through the implementation of PACS and RIS.

In recent years, hospitals have invested large financial resources in information systems, because the hospitals are increasingly compelled by legislation to reveal costs in a transparent way and to assign them to each individual department. One reason for this is from the plenary of the adopted the Rules of Procedure of the German Federal Joint Committee, which came into force in July 2005. Since 2005 the relatively strict criteria of evidence for the evaluation of methods and services have been valid for the contractual provision of care and also for acute in-patient care (Busseand Riesberg 2005).

The largest investments were made following the requirements of legislation, such as the lump-sum compensation act in billing for services from health professionals to the Diagnosis Related Groups (DRG) in the field of information systems in support of hospital administration. The modernization and integration of medical systems was often given too little importance. As a primary result of this development lower-cost solutions for individual areas have often been purchased, which, due to lack of data and process integration, have lead higher long-term costs, and, as a result, may adversely affect the competitive situation. Schweiger also emphasizes in this regard that through too short-termist management, HIS tends to be integrated poorly with information technology (Schweiger 2000).

Other weaknesses of today's systems are the cumbersome handling of different medical systems and the lack of possibilities for data exchange with other systems through incompatible data formats. The first medical information systems were stand-alone solutions and allowed no or a technically difficult data exchange with other systems. Today's systems provide the data exchange via interfaces, but although there are now standards for the design of interfaces such as DICOM and PACS for Health Level 7 (HL7) for RIS and HIS, the technical connection between two systems is a technically meticulous and detailed work at the byte level (see Chapter 2.2). Current systems are still far away from being holistic and fully integrated solutions. The entry of medical findings and actions via the computer is also limited. For example, the search for reports and frequent requests by location and content, for which doctors and nurses have to invest a high proportion of their working time, shows the degree of inefficiency.

The challenge lies not in the design of individual information systems and singular solutions, but in the integrated view of the whole complex, the whole socio-technical system, the influencing factors and the implementation of appropriate information systems. In particular, the use of information systems in the future should not be stand-alone solutions of software and hardware for the storing and processing of structured data, but as a subsystem of an integrated organization with emphasis on all of the information-processing processes, involving both human, and technical mechanisms (Busse and Riesberg 2005, Winter, et al. 1998).

Owing to the massive pressure for change in the established deregulation there has been a lot of short-termist optimization within hospitals. Unilaterally influenced fashion trends are often pursued, in which strategic policy questions are only considered only in passing (Sterman and Wittenberg 2000). The previous picture shows the need for a networked view of all factors in the implementation of information systems. In the design of teleradiology networks especially, the digitalization of radiology, together with the digitization of the entire field of medicine and the systems should be considered.

The vision lies in the standardization of structures and processes of all participants in a teleradiology network, without limiting the doctors in their freedom to take action. The structures, processes and information technology should be based on the patient’s needs and, as a flexible tool, offer the physician a platform to focus on the individual case and the patient's recovery. The physician’s approach is based on medical aspects and is not to be hindered by artificial restrictions, processes requirements, or administrative and information technology hurdles. In a teleradiology network, the doctor may, before or during treatment, quickly access all of the patient‘s relevant data with their radiological reports, without tedious and sometimes futile searches for patient data in the archives. Radiological images from past treatments can also be crucial for a current diagnosis, for example, to examine the progression of disease over a longer period or to diagnose changes. Data access in teleradiology networks is done regardless of time and place through digital images. The transport of the image data is carried over data networks, for example, over long distances by using the Internet. Access to the various systems takes place via workstations and terminals of the various systems in doctors' offices, in the various departments of hospitals, pharmacies and health insurance companies. The participating doctors and nurses can simultaneously access various patient data and evaluate them together for a quick diagnosis, report and communicate, enhancing the quality of treatment. The digital communication, storage and archiving of image data and of the patients‘ reports takes place without technical control of the process by the medical staff. This allows the physician to concentrate on the medical task and be relieved of the additional work load of the data control as much as possible. The information technology based processes in hospitals have developed progressively over many decades and were rarely questioned and often only slightly adapted to modern needs. In particular, where the interfaces between the application systems used (RIS, LIS, PACS, etc.) are concerned, stand-alone solutions were also usual (Niemann et al. 2002). This way different subsystems of HIS work redundantly on the same processes. A suitable interface architecture can help prevent this. Through new strategies in the recent research solutions strategies such as coupling, Service-Oriented Architecture (SOA), Enterprise Architecture Integration (EAI), Enterprise Process Architecture (EPA) could be integrated into HIS practice (see . Chapter 2.11).

The political power structures and the particular complexity of the medical and economic processes in the hospital have hampered optimization based change. In addition, the need on the part of hospital management and the hospital operators had not been appreciated as the economic pressure has been missing up until 15 years ago. Before implementing new information systems, which are oriented to the processes in hospitals and teleradiology networks, the processes and structures need to be adjusted to meet the current and future requirements.

An aggravating factor for the comprehensive analysis and optimization of a hospital teleradiology network are the frequent health care reforms which tie hospitals and doctors to restructuring. The previously mentioned legislation has been one such contributor. According to Hey and Huber a new reimbursement system for hospitals is forcing them to think about IT-based process efficiency (Hey and Huber 2006).

The considerable cost-cutting measures demand a high degree of efficiency whilst further increasing quality and care. Hospitals, health insurers, legislators, but also the industrial companies participating in health care are faced with the task of acting masterfully in an increasingly complex environment. The systematic support of service processes can only be performed by an information system, such as a hospital or radiology information system, if the organizational structures are adapted to the future needs of health care. With today's approaches, such as the introduction of electronic patient records and health cards, we are at the beginning of the overarching approach. This requires interdisciplinary communication, so that the designs of information systems suit medical needs and not solely the representations of theoretically oriented approaches. The success of the introduction of DRG, coding systems and future processes or systems is dependent on the ability of integration into the structures, processes and information systems in the hospital.

The provision of modern information technologies and their link in teleradiology networks is a strategic tool in view of a patient-oriented health care and the basis for improved management and scientific research in an environment where cost pressures and reform-led restructuring are so prevalent. The question arises from the digitization of radiology and the theoretical analysis of its impact on patients, processes and economics, what specific requirements for a holistic, integrated, process and patient oriented information system provided in a teleradiology network should be and how the implementation method must be designed.

The method of implementation of holistic and patient-oriented information systems in teleradiology networks integrates all factors on the overall system and provides the results in relation to each other. The integration of patient orientation is an important component that is to be implemented methodically and substantially. It is the application of the method which allows us to cope with the often underestimated complexity.

Current implementation methods are not very patient-oriented, yet they methodically integrate the relevant areas such as organizational structure, work-flow organization, quality assurance, process-supporting information technology, economic aspects or other factors. Niemann et al.also refer to this.In terms of application integration, they criticize the existing interfaces between the subsystems are not sufficiently suitable for the required business processes (Niemann et al. 2002).For this reason, the relevant integration approaches in the context of the implementation methodology are addressed (see Chapters 5.1 and 5.2).

The implementation method goes beyond the technical aspects of the implementation of information systems.It includes aspects of strategic management and is a strategic, tactical and operational tool at the same time.

1.2 Aims and Objectives

The objective of this thesis is divided into two interrelated goals, from which partial and sub-goals are derived. The first objective is to describe the digitization of radiology and its impact on patients, processes, and the economy. The digitization of radiology integrates the scientific disciplines of medicine with those of computer science. But other areas have been involved through the digitization of radiology that should also be examined in the context of the theoretical analysis. The separate analysis of patients, processes, and the economy is aimed at the categorical classification of the various factors influencing the digitization of radiology. The theoretical analysis leads to the elements that should be included in a holistic and patient-oriented method, and thus forms the basis for the modeling of the implementation method.

The second aim and the actual result part of this work is the development or modeling of a holistic and patient-oriented method for the implementation of information systems in teleradiology networks. The goal is to address the above problems and to develop appropriate solutions to an integrative approach. The implementation methodology is a theoretical approach, without its own data collection.

From the two main objectives of the work following partial and sub-aims can be derived:

- A description of the current state of scientific research and a description of the information technology used in radiology.
- A view of the issue from various different scientific perspectives and the integration of results in the method to be developed.
- In the theoretical analysis, in addition to the scientific and technical capabilities of the individual areas, their current situation is also to be described. In the practical application scientific knowledge, influenced by external circumstances is often not used or only used in a modified form.
- Coordination of the scientific requirements with the needs of the practice.
- Integration of quality assurance required by law in the theoretical analysis and the implementation method to be developed.
- The method to be developed should be open to future requirements, such as requirements for legal health care reforms and changes in its structure.
- The requirement for the method is the mapping of the entire complexity of a hospital as a representative involved in a health care system teleradiology network organization.
- Another requirement of the method is the integration of the information technology, with the wide range of instruments used, in an improvement of structural and process organizational problems in radiology, hospital and health care. The problem lies in harmonious use of the instruments, so as to meet all needs in a complex and ever-changing environment.

The focus points of the theoretical analysis and the implementation method lie in the consideration of a hospital as part of the health care system that is to be integrated into a teleradiology network. In principle, the implementation method for information systems in teleradiology networks is applied regardless of the type of organization and the organizational or business size of the organization. The amortization rate must be taken into account. The larger the organization, the more the amortization rate increases.

1.3 Work Procedure

illustration not visible in this excerpt

The thesis is divided into a total of 7 chapters, of which there are seven main chapters (see Figure 1) followed by administrative sections.In each chapter, after outlining the problem and the motivation of the work, there is a description of the aim of the work.

Figure1: Work Procedure

Introduction with statement of the problem, motivation, procedure and aims. Chapter 1

State of the Art: Systems and processes of information technology. Chapter 2

Material and methods. Chapter 3.

Theoretical analysis. Effect on: Patients, Processes, Economics. Chapter 4

Modeling of the implementation method. Modeling a method for the implementation of a holistic and patient oriented information system in teleradiology networks. Chapter 5

Discussion and outlook. Chapter 6

Summary. Chapter 7

The current situation and the state of scientific research on systems and processes in information technology are described in Chapter 2, "State of the Art". The focus of the chapter is on systems and procedures used in a hospital and the information technology used in radiology.

In Chapter 3, "Material and Methods" the methods used in this work, mainly those of literary research, are formulated. Since this is a theoretically-oriented piece of work which models an implementation method, there is no specific data collection.

The complex effects of the digitization of radiology on patients, processes, and economics are described in Chapter 4 "Theoretical analysis". Its content leads to the requirement for a new implementation method for information systems.

In Chapter 5, "Modeling the implementation method," the results gained from theoretical analysis are integrated in the development of a methodology for implementation of information systems in teleradiology networks.

The description of the achievement of the objectives as part of the results of this work is the subject of Chapter 6, "Discussion". In addition, the boundaries of the developed method are presented and a differentiation from other research areas is carried out. The conclusion of this work and the outlook on future developments and applications conclude the discussion.

The last main chapter, Chapter 7 „Summary“, briefly summarizes the entire work.In this chapter the results of the discussion and the conclusions of the work are summarized.

The administrative chapters contain the bibliography, the tabular appendix with table of figures, the CV, the acknowledgements and the affidavit.

2 State of the Art

illustration not visible in this excerpt

The digitization of radiology is based on the developments and findings of the medical specialist area of radio therapeutics in connection with the progress of information technology. Radiology as a science and the tenet of medical utilization of certain types of radiation in diagnosis and therapy (Pschyrembel 2004) is itself a fusion of electronics and medical knowledge. In addition to the traditional X-ray, computer tomography, magnetic resonance imaging, diagnostic ultrasound and positron emission tomography in the field of nuclear medicine, there are other imaging modalities such as scintigraphie or endoscopy. The technology and procedures of diagnostic radiology are not described in this work, but it will continue to refer to extracts from relevant literature (Ewen 2003, Goretzki 2004, Laubenbergerand Laubenberger 2006, Strotzeretal.1999). The digitization of radiology came about by the use of information technology as answers to radiological questions and more efficient management of radiology. The systems and methods of information technology in radiology are presented below in this chapter.

Figure 2: Information technology systems in radiology (infrastructure)

A digitized radiology department consists of a radiology information system (RIS), a Picture Archiving and Communication System (PACS) and, encompassing the whole of radiology and more, an overarching Hospital Information System (HIS). In combination with the electronic patient record, this approach leads to a filmless healthcare system (Huang 2004). A prerequisite for the integration of PACS is the introduction and acceptance of standards such as the DICOM standard for the transmission and storage of image data or the HL-7 standard (Huang 2004) for the transfer and management of patient and treatment data.

Figure 2 shows the integration of PACS, RIS and HIS, which include many of the same functionalities and whose data flows and operations must be precisely aligned with each other. To increase the acceptance by the users and thus increase the efficiency of the system, there should be a consistent and intuitive interface available to the user. In a perfectly integrated user interface, the user knows with which underlying subsystem he is working. He does not need to know which system processes run automatically in the background between the interfaces of the subsystems.

2.1 Medical Documentation and Classification

Medical documentation and classification belongs to the subject area of information technology, as, in the same way as with as a programming language or a communications standard, the user turns the information from the real world into a code, thereby making it standardized, comparable and transferable.The information technology processing of data is carried out according to a predetermined scheme, the medical documentation and classification is such a scheme and therefore a prerequisite for the processing of data using information technology (Zaiß et al. 2005). In order to collect structured information and to be able to analyze it indexing languages are used. Each indexing language is the sum of its systematically compiled descriptors that are allowed to describe situations in a field. Each indexing language is a hierarchical classification system with upper and lower terms. The classes of a classification for that documented area are complete and disjoint.

2.1.1 ICD-10 (International Statistical Classification of Diseases)

The ICD-10 (International Statistical Classification of Diseases and Related Health Problems) was created as part of international health classifications by the World Health Organization (WHO). The German Federal Ministry of Health and Social Security commissioned the German Institute for Medical Documentation and Information (DIMDI) to translate and publish it in German (DIMDI 2005). The abbreviation ICD stands for „International Statistical Classification of Diseases and Related Health Problems". The figure 10 denotes the 10th revision of the classification. It is used in Germany as the encoding of the cause of death of and the encoding of diagnoses in outpatient and inpatient care. It is thus the basis of the official cause of death statistics. Since 01.01.2000 the ICD-10 has been used for the encoding of diagnoses in outpatient and inpatient care (§ § 295 and 301 used SGB V), particularly for the purposes of the flat-rate fee system G-DRG (German Diagnosis Related Groups). For these purposes, the ICD-10-GM is used, which was known as the ICD-10-SGB-V until 2003. GM means "German Modification", SGBV stands for "Social Law Book V". These special editions of the ICD-10 based on the German-speaking ICD-10 WHO edition, were, however, significantly altered for the purposes of the Social Law Book V. The ICD-10-GM includes the systematic index and the alphabetical index with an extensive collection of coded diagnoses from the language used in the outpatient and inpatient care.The rules for the encoding of diagnoses in inpatient care (§ 301 SGB V) are recorded in the German coding guidelines which are not from the DIMDI, but created and well maintained by the bodies of self-regulation.

2.1.2 OPS (German Procedure Classification)

The OPS (German Procedure Classification) is published by DIMDI on behalf of the Federal Ministry of Health. Currently, the 2005 version is used in out-patient applications for the encoding of measures. From April 2005 Section 5 of the OPS has been used in statutory provision of in-patient operations.

2.1.3 MeSH (Medical Subject Headings)

MeSH (Medical Subject Headings) is polyhierarchically structured and traditionally consists of three parts: MeSH Tree Structures (systematic part), MeSH Annotated Alphabetic List (alphabetical list) and Permuted MeSH (permuted list). At the end of 1980 the original NLM MeSH and the subsequent German MeSH were reorganized from a descriptor hierarchy into a concept hierarchy. The MeSH vocabulary remained unaffected. The advantage is a better picture of the relations between MeSH terms and the improved integration into the concept-based UMLS thesaurus.

2.1.4 UMLS (Unified Medical Language System)

Unified Medical Language System (UMLS) contains medical terms and semantic relationships between terms. The terms come from about 100 heterogeneous conceptual classification systems and medical nomenclature, currently in 15 languages. The UMLS was started as a project in the late eighties by the National Library of Medicine (NLM). The project aims to semantically accumulate the different conceptual systems and connect them with each other in order to help to create conceptual links between user queries and relevant technical information and heterogeneous online information systems.

2.1.5 SNOMED (Systematized Nomenclature of Medicine)

SNOMED stands for "Systematized Nomenclature of Human and Veterinary Medicine" and was developed by the College of American Pathologists (CAP). It divides medical terms used into categories. These subdivisions and their combinations allow almost all the medical terms used to be recorded. SNOMED is integrated with HL7 and DICOM. A DICOM-SNOMED micro glossary was defined for DICOM (NEMA 2005).

2.1.6 UMDNS (Universal Medical Device Nomenclature System)

The nomenclature for medical products Universal Medical Device Nomenclature System (UMDNS) was developed by the Emergency Care Research Institute (ECRI) for the encoding of medical products. DIMDI has published the German translation of the nomenclature in version 1.0 since 1996. The extended version 1.1 of the German UMDNS contains all of the complete and unaltered main German terms of version 1.0 (from 1996) and, in addition, further German synonyms for names of medical products. The added German synonyms should help users to find relevant UMDNS main terms more easily. The encoding of medical products is supported, as various conceptual entry points can lead to the main concept for a product. The nomenclature UMDNS essentially covers all medical products and products from other areas, such as hospital furniture. To support the statutory functions under the Medical Product Act (MPG), the application of a uniform nomenclature for medical products is required.

2.1.7 LOINC (Logical Observation Identifiers Names and Codes)

LOINC is an internationally recognized system for unique coding of tests, particularly in laboratories, and can be used for effective data exchange with other systems in a medical clinic or practice. This makes LOINC another important component of a functioning telematics infrastructure in Germany. DIMDI promotes the introduction of LOINC and has assumed the active role of the central data storage and exchange of information with the relevant national and international institutes, project groups and industry. The success of any telematics architecture is, among other things dependent on the ability of the electronic exchange of medical data. LOINC is a pragmatic system to uniquely identify studies. Originally created for laboratory tests by the Regenstrief Institute in the USA, LOINC was extended to include clinical and medical-technical investigations. To avoid duplication and incompatibilities with existing standards and to ensure safe public access results, information and activities related to LOINC flow together at DIMDI.

2.2 Communications standards in Radiology

Information and communication technology is increasingly penetrating outpatient, inpatient and administrative health care areas. The vendor-independent integration of information and application systems such as HIS, RIS and PACS is the central task that can only be achieved by using internationally consistent communication standards. In addition, the function processes of these systems must be harmonized to ensure temporal and logical dependencies are included. In radiology DICOM is established as the image communication standard and HL7 is leading and generally established for the transmission of management data. This shows the merging of internationally operating procedures and IT businesses in the IHE (Integrating the Healthcare Enterprise), in order to model patient-and process-oriented scenarios, based on industry standards such as DICOM and HL7 (IHE, 2006).

2.2.1 DICOM Standard

DICOM stands for Digital Imaging and Communications in Medicine and is an open and manufacturer independent standard for storage of medical images and for medical imaging devices to communicate with each other (NEMA 2005). The standard was developed in 1983, originally by the ACR (American College of Radiology) and NEMA (National Electrical Manufacturers Association). The system, adopted since 1993, DICOM 3.0, is based on the old ACR-NEMA standard 1.0/2.0. An independent panel of 26 working groups is working continuously on the further development of the standard. Changes or innovations are adopted in the so-called supplements and incorporated into the standard in a cyclical manner. Since 1995 DICOM has also been accepted as a formal standard in Europe. Since the 2003 version 3.5 DICOM is not versioned. There are only a basic standard and the ability to specify these further and add to them. DICOM consists of the parts listed in Figure 3 (NEMA 2005).

In order to integrate devices and software from different vendors into a network the radiological equipment manufacturer may use DICOM-compliant interfaces. With the declaration of conformity (Conformance Statement) the manufacturer documents which DICOM services are supported, which extensions have been implemented and how communicating with other systems functions. This makes it possible to make statements about the compatibility of devices and programs and to connect them in a heterogeneous network. DICOM supports many radiological modalities such as X-ray, CT, MRI, ultrasound,computer radiography, angiography, digital movies and video.

illustration not visible in this excerpt

Figure 3: The Parts of the DICOM standard (NEMA 2004)

A DICOM image is saved with a syntactically and semantically predefined list of data elements or attributes, where many additional pieces of information are included (Bidgood and Horii 1992). These include patient information such as name, birth date and a globally uniform identification number. Similarly, the information on modality (e.g. X-ray, CT, MRI) and recordings such as device parameters, calibration, radiation dose and contrast agent are stored, and finally, the image information, such as resolution and windowing of important data, so the image can be displayed properly. For each modality there is a DICOM requirement which the data elements must complete, which are optional and which are completed under certain conditions, for example, only when contrast agent has been used.

The DICOM standard offers many different services, therefore only the most important are mentioned below:

- DICOM image transfer service (Storage Service Class)
In the implementation of the Storage Service class a sending DICOM application can send an image to a receiving application. In addition to individual images complete radiological investigations can be transferred.
- DICOM Network Services
Before an exchange takes place between two DICOM applications a connection must be established. During this it will be negotiated which DICOM services are used and how the data is transferred (e.g., number formats and compression). A connection can only be established when the same systems are present.
- DICOM image archive service (Query / Retrieve Service Class)
This service enables a search for PACS images with selectable criteria (patient, date of inclusion, modality, etc.) and to request the selective images from the archive.
- DICOM Print Services (Print Management Service Class)

The print service allows the control of laser exposure devices or printers on a network, whereby one image setter can be used by several modalities and diagnostic work stations.

- DICOM Modality Work list service

This service provides the interface to other information systems. It takes work orders, including automatically collecting the associated patient data from the information system (such as HIS or RIS).

The exchange of storage media is made possible in the DICOM standard by defining the application profile for each DICOM media (Kimura et al. 1995). It describes which modalities‘ pictures may be stored on the disk, what number formats and compression may be used and which disk is to be used. Besides the actual DICOM images, every DICOM disk contains a file containing the key information of all images such as patient name, modality, unique identification number, in order to quickly and efficiently search for images without having to read all the information to disk.

In Germany, in 2002, teleradiology was defined in § 2 of the X-ray Ordinance for the first time and allowed under certain conditions. Had the necessary expensive specialized software not delayed the distribution of teleradiology in Germany, the use of non-proprietary transmission protocols in the standardization of telemedicine radiology may have been established at the 85th German Radiology Congress in 2004 in Wiesbaden. Supplement 54 of the DICOM standard in 2002 defined how DICOM image data in the original format could be attached to an e-mail as S/Multipurpose Internet Mail Extension can (MIME) (Engelmann et al. 2005).

The route of transmission by e-mail was chosen because ad-hoc connections could be added without having to make changes to the existing security settings HIS internal firewalls (Engelmann et al. 2005). Moreover, as the DICOM standard is integrated in the Windows operating system both partners are no longer required to use the proprietary teleradiology software of the same manufacturer. Consequently, this safe and cost effective data transmission method has facilitated the spread of teleradiology since the introduction in Germany. DICOM emails make both teleconsultations possible not only for all of the applications mentioned in the X-ray Ordinance, but also emergency care or the weekend and on-call shifts, as communication links with the diagnosis. The S / MIME encryption prevents unauthorized access to the supplied data by third parties and fulfills the requirements for data protection (Bartoll and Hoffmann 2006).

2.2.2 HL7-Standard

The communication standard HL7 (Health Level Seven) (Heitmann 2006, Hinchley 2003, Heitmann et al. 1999, Hammond 1993) is the standard for the exchange of data between information systems in the administration and medical applications. HL7 was developed in 1987 in the United States by users and developers of hospital information systems. User groups in many countries continue to develop HL7 and adapt it to the country-specific requirements of the respective health care systems.

HL7 standardized the structure and format of messages on the top layer of the ISO / OSI reference model (International Standardization Organization / Open System Interconnect) (Heitmann, 2006), the application level. The model consists of a total of seven layers, with the lowest level represented as the physical connection, such as a cable connection. Thus, the platform independence of HL7 is clear, for the technical implementation of a connection is irrelevant to the message format.

illustration not visible in this excerpt

Figure 4: Schematic of a HL7 message (according to Heitmann et al. 2006)

The transferred data is stored in a structured form in variable length fields, which are divided into segments. Different HL7 segments are defined for different tasks, and can be assembled on a modular basis in accordance with the requirements of customized messages. The various messages are in turn grouped in non-disjoint groups. A segment can be found in various pieces of news and groups. The HL7 segment includes groups for applications such as:

- Admission, transfer, discharge
- Ordering goods and services
- Laboratory results
- Medicines
- Patient management
- Billing
- Patient records

The current version of HL7 is version 3 with the feature of an object oriented information model and the requirements for the electronic health record (Heitmann 2006).

2.2.3 IHE Initiative

IHE (Integrating the Healthcare Enterprise) (Huang, 2004, IHE 2006), was started in 1998 and is an American initiative of the RSNA (Radiological Society of North America) and the association of providers of medical information - HIMSS (Healthcare Information and Management Systems Society). It has evolved into an international initiative with consideration of requirements from Europe and Japan. The promoters of the IHE in Germany is the German Roentgen Society (Deutsche Röntgengesellschaft) and the Association of Electronic Medical Technology Industry in the ZVEI (Fachverband Elektromedizinische Technik im ZVEI) (Eichelberg et al. 2004). The goal is to support a seamless exchange of information of digital systems in health care through cooperative and focused action by the user and manufacturer. A cornerstone of the initiative is the technical framework in which the most important IT components of a health care establishment are identified and the interaction between the components through transaction on the basis of international standards such as DICOM and HL7 are described (Eichelberg et al. 2004, Krüger-Brand 2004, Wine 2003).

The modeling of typical application scenarios in IHE integration profiles (see Figure 5) enables users in hospitals and medical practices without knowledge of DICOM and HL7, to make statements about the integrational ability of application systems. Above all, integration profiles and the participants involved are defined. Through the specification of in participants association with the integration profiles the transactions of the IHE are defined.

illustration not visible in this excerpt

Figure 5: IHE integration profiles (according to IHE, 2006)


SCIPHOX (Standardization of Communication between Information Systems in Physician Offices and Hospitals using XML) is a German project, which began in 2000. It is supported by many organizations such as the HL7 user group in Germany, the Qualitätsrings Medizinischer Software (QMS), various federations and associations, some universities and manufacturers of hospital and practice information systems. This involves the specification of definitions for communications tasks of IT systems in private practices and hospitals (Heitmann 2003). In the first phase the substantive requirements of communication between outpatient and inpatient care facilities were specified. The basic idea lies in the definition of electronic documents and forms based on international standards with implementation through Extensible Markup Language (XML).

The technological basis specification for the SCIPHOX project is the Clinical Document Architecture (CDA), which was developed within the HL7 group (Dolin et al. 2001). CDA has replaced the previously popular xDT standards (such as billing storage media, treatment storage media and laboratory storage media) of the Federal Association of Statutory Health Insurance, for the exchange of data in the field of ambulatory medical care (Schnettler and Fuhrmann 1997). The CDA documents are part of an electronic patient record.

Documents created in SCIPHOX are information objects that are exchanged based on the real situation between the systems and which persist in the receiving system as a whole and thus remain whole over the time. CDA documents offer the possibility of a standardized structured transfer of clinical information. The information to document, event, patients are stored in the "header" and the clinical data such as diagnoses, medications, therapies, information on reintroduction, proposed appointment dates are stored in the "body".

2.3 Medical Imaging

Medical image processing is the key technology for modern image-guided diagnosis and surgical support. The best display of an image, with respect to a specified question, is achieved by medical imaging. It is the prerequisite for an object-specific perception, observation, medical diagnosis and integrated assessment. Various demands are put on the brightness, contrast, processing, number and size of the image elements and the format of the displayed image (ZVEI 2005). There are different methods available to delineate and evaluate structures in images. One requirement for the imaging process is the digital delivery of images is of the modalities. At the beginning of the imaging process, the images are processed and, for example, put through a noise filter to eliminate interference and to ensure the quality of the final result is not affected from the very start. After that, the definition of the depicted structures and extraction of significant features from the image takes place (Park and Keller 2001). The characteristics of the image are quantified and classified, whereby a classification of the image information displayed in a semantic context is made possible. The assessment of the information takes place in a final analysis. The demarcation of the pixels of objects and feature extraction are the main steps of image processing. The global and local feature extraction is differentiated. The global feature extraction does not perform object detection and relates the properties used to the entire image, as is the case with histogram information.

The content categorization of images in the global feature extraction is only possible if the pictured object dominates the picture (Keysers et al. 2003). In the local feature extraction only a subset of the image is studied, whereby specific properties of the structure in the image area shown are extracted. A survey of the illustrated of structures becomes possible through the localized features (Kauffmann et al. 1998). In addition, statements can be made about the nature of the investigated structures. The statements made on the extension, size or volume information of tissues and organs is of great importance in medical diagnosis. Safe findings on the physiological and pathological changes can only be drawn through such statements.

Prior to the extraction of local object-specific features thesegmentationof relevant structures of the content related regions takes place. In segmentation, adjacent pixels, which meet certain homogeneity criteria are processed by mathematical methods of image processing and are displayed in an emphasized manner (Lehmann et al. 1997). Most segmentation methods are available as a special solution to certain limited domains (Schreckenberg et al. 2000). The knowledge-based portion of these procedures is often based on heuristics, whose algorithmic structure is tailored only to a specific problem (Kauffmann et al. 1997). The following segmentation types are primarily distinguished (Lehmann et al. 1997, Schreckenberg et al. 2000, Park and Keller 2001, Keyser et al. 2003, Huang 2004, Dreyer and Kalra 2005, LaubenbergerandLaubenberger 2006):

- Point Oriented Method
Through simple threshold operations the foreground is shown separately.
- Edge Oriented Method

In the segmentation process by the edge oriented method boundary edges or contours in areas can be found. An edge is a boundary between two regions with certain levels of grayness. These regions are largely homogeneous, so the transition between the two regions based on the change in grey levels can be calculated. Most methods for edge detection based on a calculation of the gradient between adjacent gray values of an image. The greater the magnitude of the gradient, the more pronounced the edge.

- Region-based Method

Components are the maximum continuous parts of the picture. The Region Growing Method, the components can be determined. The segmentation is based on one or more pre-determined starting points or seed cells. If two areas touch over the course of the algorithm the homogeneity criterion will be reviewed, and the areas united if conformity is found. Another approach is the split and merge algorithm. The image is recursively decomposed into sub-areas until each sub-region meets the homogeneity criterion. In the merge phase sub domains which meet the same homogeneity criterion and are adjacent are grouped in one area.

- Rule Based Method

A precondition for rule-based methods is the knowledge of the segment forms before the procedure. The image is compared with a specific pattern or given Hough transformation vectors to find straight lines.

With thewindow technique, the gray values of the original image are transformed on a diagnostically meaningful gray value area. The window shows the selected gray value area of pixels. Especially for images with large contrast differences subtle differences in contrast can be represented and optimized.

Thedetail enlargementor zooming allows the representation of individual image sections on a larger scale without increasing the matrix. Small image details can be seen more easily without changing the image (LaubenbergerandLaubenberger 2006).

2.4 Image Compression

The storage and distribution of images is a central task of PACS systems. Depending on the investigations and recording type the quality requirements vary with the images produced. For example, in ultrasound images of a resolution of 320 x 240 or 640 x 480 pixels at a gray value resolution of 8 bits per pixel are sufficient. Other the hand, in digital X-ray images a resolution of 2048 x 2048 pixels are required and a gray value resolution of 10 bits per pixel. The size of medical images grows progressively with increasing image and gray value resolution of imaging devices. The total data volume of radiology in the average hospital is therefore very large and makes heavy demands on storage capacity, performance and data transfer. The effective storage and fast transmission of medical images between the transmitter and receiver with direct phone lines, local or global networks like the Internet are important strategic issues of digital radiology.

High data rates are usually only available in local area networks (LAN - Local Area Network), whereas the communication outside the immediate proximity is only possible through networks with limited capacity such as ISDN, GSM or the Internet. To meet these demands, the picture data must be compressed.

illustration not visible in this excerpt

Figure 6: Image compression algorithms

For the compression of image data algorithms are used which remove data from the original data record. These can be grouped by type of reconstructibility in lossless and lossy algorithms. The ratio of the file size of the original image to the compressed image is reproduced with the compression rate. The lossless algorithms allow a complete reconstruction of the original image data; however, the maximum compression rate is only at 1: 3 (Gillespy and Rowberg 1994). Lossy image compression algorithms lose information from the original image, which can achieve compression rates of up to 1:1000. In general, the higher the compression rate is the lower the quality of the reconstructed image. Most algorithms allow the user to select the compression rate.

In medicine, the demands on image quality are very high, as the analysis of the images will lead to a diagnosis from which the treatment of the patient is derived. A misinterpretation due to missing or incorrect information, such as in the case of a tumor, can have a dire impact on the health of the patient.

Therefore, lossy image compression algorithms are only applicable in the field of diagnostic medicine if the type of information being lost would have no perceptible impact on the quality. The quality of the reconstructed image is determined by the difference between the original image and the reconstructed image. The higher the quality of the reconstructed image the less difference there is between images.

How strongly a digital image will allow for compression without loss of information depends largely on the image content. This means that for an image with different compression algorithms, different compression ratios can be used. Thus the problem to be solved is to find, for each image, an automatic adaptive selection of an appropriate compression algorithm based on a preliminary analysis of the image content and the requirement specification.

Lossless compression enables an exact restoration of the original data (Gillespy and Rowberg 1994). The principle is based on redundancy reduction. Here statistical signal properties are exploited and predictable signal components are not transferred.

Lossy compression eliminates irrelevant information and therefore the image cannot be entirely reconstructed. Human perception is such that the eliminated information is not noticed by the viewer. The quality of the data is not compromised. The danger lies in assessing what information is relevant and which are not. The loss of a few bits of color information alone can lead to a false diagnosis.

Many procedures involve the steps of signal transformation, quantization and entropy coding. The first two steps reduce the entropy increase of the input data and thus the efficiency of entropy encoding. The actual compression takes place with the quantization step.

2.5 Picture Archiving and Communication System (PACS)

2.5.1 Historical Development of PACS

The beginning of computer-aided medical imaging was in the 1970s. In 1972 K.H. Höhne began the ISAAC project (Interactive Scintigram Recording and Analysis with a Computer) for collecting, storing and processing of scintigraphic data (Höhne et al. 1974). He established a research project for evaluating angiographic X-ray series (Nicolae et al. 1979) with the necessary hardware requirements for the acquisition and processing of medical images.

The management of large volumes of data, such as those resulting from medical imaging, was shown by M. Böhm and KH Höhne in 1979 using data reduction and inverse transformation (Böhm et al. 1979). Lemke described the 1979 applications of these techniques in the project "COMPACT" at the Technical University of Berlin. It had already allowed the three-dimensional representation of brain structures and an automated calculation of ventricular volume by cranial computed tomography (Lemke and Stiehl 1979).

The term PACS (Picture Archiving and Communication System) is based on one of the concepts formulated by H. U. Lemke (Lemke 1979). The elements of a PACS are modules with functionalities for processing of incoming image and patient data for the transmission or communication of data to other systems and for compressing and archiving of the amount of data (see Figure 2).

2.5.2 The PACS Communication Concept

The communication concept of a PACS is based on the support of the DICOM protocol and its implementation. The implementation is left to the underlying data processing concept. Modern systems are based primarily on the object-oriented technology such as Common Object Request Broker (CORBA), the Java client / server and Web technology (see Chapter 2.11).

In an object-based architecture different imaging modalities that have a DICOM interface integrate better. Another factor in system selection is the platform independent object-oriented architectures (Heuer et al. 1999, Klaiber et al. 1999)

Modern PACS are based on Internet technology. Thus different standard tools of this technology such as browsers, e-mail clients, plug-ins, etc. act as a user interface and offer a simple, fast and secure access to multimedia archives. The location of the data is not important. The DICOM data packets are translated into the standardized Internet Protocol TCP / IP (Transmission Control Protocol / Internet Protocol) and carried on a local or global network (see Figure 2) and translated back into DICOM data packets in the PACS of the receiver (see Chapter 2.2.1).

The realization of the security concept, such as through a firewall, protects the data and the entire system against data loss from viruses and unauthorized access.Should the image data from the modalities or external systems not be available in DICOM format, image reconstruction takes place is the other picture format in the DICOM format. Knowledge of the semantics or the existence of the translation table of the other image format is a requirement. In a PACS the images from the modalities are available regardless of time or location. The doctors view the pictures on a certain kind of screen chosen for that purpose. Pictures of past studies can be retrieved within seconds from a networked central archive. The time taken for diagnosis is reduced by the direct access to stored images from a diagnostic work place.

This is confirmed, among others, by Osada et al. in the framework of a research report on HIS integration based on EAI at the medical institutions of the Westfälische Wilhelms-Universität Münster. The prompt printing, on a HL7 basis, of findings from the Central Laboratory on all 25 printers in the stations and departments was very positively assessed, as it relieved many doctors of requesting findings by telephone (Osada et al. 1998). The work process is accelerated by PACS and reduces the burden on patients during investigations. In principle, X-ray films can technically - apart from through legal reasons and reasons of habit - be ever more replaced, which also reduces the variable costs of hospitals and medical practices and productivity can also be increased further.

X-ray films are unique documents which must be kept in storage 30 years and take up a lot of space. Studies show a reduction of diagnosis times at optimized diagnostics stations with a PACS over conventional film diagnosis (Bick 2000). The omission of the film development time speeds up the diagnosis and also the high image quality, with the possibility of magnification of an image, increase diagnostic certainty. The quality of the further care patients receive is also supported.

2.5.3 Modern PACS and further Development Trends

Earlier PACS were created as stand-alone systems without connections to other systems. Modern PACS are integrated with other systems, such as hospital information systems and radiology information systems. The patient data is retrieved from the HIS or from the integrated electronic patient record therein and associated with the images of the various modalities in the PACS. The radiological test data is retrieved from the RIS and also linked with images in the PACS.

The description of the realization of a PACS in German radiology clinics and institutes and their methods and advantages are described in some publications (Fründ et al. 2001, Heiss et al. 2000). Especially for diagnostic radiology the realization of PACS is of ever increasing importance (Weber, 2004, Baumann and Gell 2000, Henri et al. 2000, Hruby 1995).

But, in the case of digital radiography (DR), it is also the economic considerations arising from the results which are delivered by the vendor independent DICOM standard of communication (see Section 2.2.1) for medical images and image related information which can make the introduction of a fully integrated PACS advisable.

The expansion of Positron Emission Tomography (PET), far beyond purely research-related questions, leads to increased image storage requirements. The combination of PET images at the cellular level with anatomical CT images in order to be able to identify lesions is one important future application, which is dependent, given the high data rate, on the disk space of the cooperating institutes. Similarly, the proposed increased use of image generating procedures in mammography and cardiology, with the associated increase in integration and communication needs, further increases the quantity of data to be stored (Bon Secours Health System, 2001).

The PACS based DICOM communication has already been discussed in the chapter on DICOM email (see Chapter 2.2.1). The benefits of progressive integration are evident both in connection with radiology related RIS functionality, as well as in transcription and diction using voice recognition. Access to patient information and radiology images, as well as communication with colleagues or the dictation of current case history reports are made possible from a „single point of entry" (Bon Secours Healthcare System, 2001).

Through the Application Service Provider (ASP) system it is possible for hospitals and doctors to outsource radiological image data archiving to third parties using Web-based technologies. Patient related "work lists" can be called up on demand and sent through new compression algorithms to the radiologist and the hospital departments involved. Hence the radiologist is allowed increased mobility, as they also have mobile access to relevant patient data and can communicate with colleagues via secure DICOM-e-mail connection (Bon Secours Health System 2001).

PACS distributors can be classified into the following categories (Bon Secours Health System 2001):

- Service provider for imaging radiological procedures
- Service provider for film technology for information systems
- Service providers for special solutions
- Application Service Provider (ASP)

Important service providers for radiological imaging procedures are companies such as General Electric (GE), Siemens and Philips. The large photography companies Agfa Corporation, Eastman Kodak, and Fuji Medical also offer PACS for video. For example, through the acquisition of ADAC and Dynamic Health Care Technologies the company Cerner has focused on PACS information systems. ALI Technologies and DR Systems are two examples of companies that focus on the supply of PACS special solutions.

Finally, the companies Emageon and Imphact have positioned themselves successfully as application service providers. They offer companies with financial preferences the required PACS storage space on demand, and thereby in a flexible and cost-effective manner (Bon Secours Health System 2001).

Depending on requirements and areas of application of telemedicinal oriented medical institutions there are a wide variety of PACS implementation options to choose from. When taking the decision to implement a PACS solution the benefits gained from that PACS for different departments have to be weighed up with the concrete requirements and the financial framework.

The acceleration of diagnosis and decision making represents a benefit resulting from the fast simultaneous access to patient image data and files for the participating doctors. On the other hand, the introduction of PACS in a radiology department leads to productivity gains compared to pure film-based, serial information systems. This is due to those with PACS also being given due accessibility to previous investigations and information and the consistent image quality, which helps the radiologist to a more efficient interpretation of the study images (Bon Secours Health System, 2001).

2.6 Hospital Information system (HIS)

In the operation of a hospital a number of different people (including doctors, nurses and administrators) work in various different areas (e.g. radiology, ambulance, ward, laboratory, administration).The different areas and people constantly exchange information.Therefore, in a hospital, as is also the case in other companies, information processing takes a very high priority (Winter 1997).

A hospital information system (HIS) is the entire information processing and information storage subsystem of a hospital, whereby it is not just about computer systems and networks and the computer-based application systems that are installed on them, but it is about the information in a hospital as a whole (Haul et al 1996). To address this complex problem, there are a large number of interconnected and communicating subsystems (see Figure 7).

These can be grouped according to function in two groups, the clinical information system with all of the medical functionality and the hospital management system for managing the administrative, managerial and technical supply functions (Hannan 1991). Belonging to the clinical information systems are, depending on how they are viewed, the department systems such as the radiology information system for radiology. Due to the complexity and partly redundant functionality of these systems, such as patient management, a separate consideration is appropriate (see Section 2.7).

The use of electronic patient records is also seen as highly significant. In scientific literature, the electronic patient record is counted by Hannan as belonging to clinical information systems. However, it seems useful to define a new group for the electronic patient record and other systems that are used by both clinical and administrative systems.

Therefore, the central systems which include global functions, which both the clinical information system and the hospital management access, are to be considered as a third group. The electronic patient record system is an important part of central system. It is a digital collection of documents and information to map and document the medical care of patients (see Chapter 2.7).

illustration not visible in this excerpt

Figure 7: Components of a Hospital Information System

In Germany, today, there are about 70 vendors of hospital information systems, which offer the complex and wide-ranging functions of very heterogeneous hardware platforms and operating systems (Krüger-Brand, 2004). On account of this, as early as 1991 an international working group published various HIS quality criteria (Ball et al. 1991). All information processes, including PACS, as well as the management for administration and maintenance should be supported by a hospital information system. This allows workflows to be supported and optimized.

The HIS is an interface between the applications in the hospital and external applications of third parties outside the hospital, such as other hospitals or health insurance companies. In particular, the support of clinical applications and procedures in the hospital is important for a HIS. This way HIS has a central and strategic importance for a hospital or a practice.

The main tasks of the HIS include the timely and context related provision of current patient data for an authorized group of people. The important thing is the appropriate form of the presentation of the information. Medical knowledge about diseases and drugs is also made available by a HIS. Information on the quality of patient care and the relationship between costs and benefits for hospital services are further responsibilities of a HIS. Further descriptions of the duties of a HIS and the new impetus on useable applications can be found in numerous scientific publications (Grätzel vonGrätzin 2004, Huang 2004, Trill 2002, BraunvonReinersdorff 2002, Winter AI et al. 1999 Winter AF et al. 1998).

In contrast, Endsleff stresses the importance of the factor of communication and its integrity: "The more recent evolvement has been characterized by the emergence of message based integration engines and new versions of legacy systems, including hospital information system (HIS), which attempts a higher degree of openness by being able to communicate messages "(Endsleff et al. 2002).

The Center for Innovative Health Technology names the optimization of business processes in the HIS as the key success factor for the integration of digital data into hospital facilities and equipment or interdepartmental treatment processes. This is primarily about uniting the HIS areas of patient management, operations management, financial management, computing, as well as testing and certification (Hey and Huber 2006).

According to Zielinski, depending on the terminological alignment, radiology information systems (RIS) used for electronic processing, analysis, storage and recall of the film recordings can be also be counted. Specialized subsystems are constantly providing more services electronically. Also, the PACS is an example of something which processes the medical images of different modalities (X-ray, computer tomography, magnetic resonance imaging, ultrasound, and Nuclear Magnetic Resonance). These can then be interpreted according to need (Zielinski et al. 2006). Zielinski et al. summarize the most important tasks in HIS summarized as follows (Zielinski et al. 2006):

- Storage and monitoring of patient’s condition:

A. Accurate and electronically stored medical records of patients (eg drug allergies) are provided
B. Visual and auditory warning systems are generated in the event of abnormal test results or other important data
C. Time intervals and / or testing periods for tests on patients to be specified
D. Data Processing (DP) and analysis for statistical purposes and research oriented purposes

- Management and Data Flow:

E. Support automated patient data transfers between departments and institutions
F. Enable graphic or digitized diagnostic images from the hospital database based on the integrated retrieval system
G. Digital signatures, in order to create internal orders electronically
H. Communication by Laboratory Information System (LIS)
I. Registration of human resources and their properties

- Financial Aspects:

J. Efficient administration of finances
K. Use and monitoring of medicines and effectivity of the ordering process
L. Expected and actual treatment costs are listed and reported
M. Automated representation of the needs of the nursing staff
N. Status analysis of bed occupancy and overall performance in the HIS

The relevance, completeness and timeliness of the information set quality criteria for the evaluation of the HIS. At the same time it applies that all patient information and that of the administrative processes for all activities are to be available in the HIS. Labor requirements, findings, tools to support the therapeutic and diagnostic process and the medical histories of patients are included, as well as search functions, or the documentation or the creation of letters (Zielinski et al. 2006).

The linking of systems through interfaces and the integration of the subsystems are considered strategic and, at the same time, critical success factors for a hospital information system. This trend is of moving positively toward standardization in the area of the interfaces of the systems in the health system such as DICOM or HL7 (cf. Chapter 2.2).

2.7 Electronic Patient Files

The conventional paper file has a long history and offers some advantages over electronic types. It is handy and can be easily transported. By the addition of more forms it can be easily and inexpensively extended. It can be read directly by the medical staff without technical assistance, special training or instruction. Doctors and nurses are experienced in dealing with paper files and have adapted their working practices accordingly. The disadvantages of the conventional paper file are significant and outweigh the advantages. It can only be available in one place at one time, losable, non-sortable or filterable, not strongly standardized, not automatically evaluated and a selective search of data is not possible. Since the comprehensive collection of all data belonging to a patient is not supported the optimal medical care of the patient is not ensured and this can result in multiple investigations or inconsistent data. Through electronic documentation and an electronic patient record the disadvantages of the conventional medical record can be avoided, with the general aims and principles of medical documentation remaining valid (Schmücker 2003).

The electronic patient record is, technically speaking, a collection of data in a database, which is managed by application programs. It is a key part of hospital information systems. All relevant data and the investigations and interventions for one patient are collected in the electronic patient record. The data is stored on digital media and are always available electronically. The electronic patient file contains the following information (Schmücker 2003):

- Personal details (Name, Address)
- Billing information
- Case or medical history
- Clinical test results
- Diagnoses
- Therapy
- All digital pictures from various modalities
- Pictorial archiving of historical pathologic findings
- Important treatment data and results of control tests
- Nursing measures

The electronic patient record allows the patient and study data to be stored without redundancy, in a complete and structured manner. The level of structure of the data allows the primary importance of information and context to be maintained. The complete storage of sensitive patient data requires highly reliable security in relation to the access protection and the long-term archiving of the data.

The acceptance of electronic patient records depends on the speed of access from the data. This includes not only the technical and logistical access but the user-friendly facility to search for information. Also, simultaneous access to a patient's data is made possible by the electronic patient record. The central data storage for any possible representation of patient data increases the transparency of medical documentation. The provision of relevant data allows the physician to assess and prioritize information better and to make decisions more transparently.

Conceptually, the electronic patient record is mature and highly developed. In daily practice in hospitals the use of the electronic patient record is still in its infancy. This is not a technical problem, but the result of the inhomogeneous hospital information systems market with very complex application interfaces on the one hand and a partial lack of acceptance by the other doctors on the other, which, among others, may be caused by errors in the implementation project (see Chapter 5).

Digital documents, such as electronic patient records are not recognized under the Judicial Code as certificates and may be evaluated as evidence if a judge sees fit. The evidentiary qualification of the digital document is therefore a procedural risk. In the absence of analog images or paper files the legal recognition of the digital archive must be ensured, at least through the application of system security, propriety and auditability. Through following the appropriate advice the resulting legal disadvantage in a legal dispute may be averted (Schmücker 2003).

2.8 Radiology information systems (RIS)

The Radiology Information System (RIS) is an information system, which covers the special requirements of radiology. Both administrative and clinical processes are supported by the RIS (Huang 2004). The entire process, from patient registration, to scheduling, the organizational planning of the department, the preparation of reports, diagnosis coding, management of the photographic material, to the documentation and communication of all relevant patient data is covered by a RIS. The RIS does the following tasks (Huang 2004):

- Making the investigation requirements request from the HIS with scheduling
- Schedules for patients, equipment and rooms
- Planning transport of patients
- Provision of labels
- Transfer of patient data to the radiology equipment (Work list Management) with data acquisition for digital images and their transmission and archiving
- Maintains the statutory X-ray book
- Documentation of material consumption
- Supporting the findings
- Digital diagnosis via a networked dictation system, automated distribution of digital dictation to writing work stations on the network
- Integration of speech recognition systems
- Report letters with standard text
- Transmission of discharge summaries via email, intranet or fax
- Administration of the X-ray archive
- Connection to a PACS for the processing and transmission of medical images
- Activity recording and performance encoding
- Transfer of the service costs for billing
- Data transfer to electronic patient medical file for archiving
- Statistics and analysis

The roles and functions of RIS overlap partially with the HIS and precise coordination of the interfaces during the implementation of systems is required in order to achieve redundancy-free integration. The standard communication systems described in Chapter 2.2, HL7 and DICOM, are used. The following figure shows a possible scenario according to Kirn:

illustration not visible in this excerpt

Figure 8: Communication between the (tele-) radiology information systems and PACS (by Kirn, 2007, p.6)

New developments in some areas of radiology have led to increased memory requirements in PACS. The computer-based positron emission tomography (PET), especially the many layered process of magnetic resonance imaging (MRI) and computer tomography (CT) may each require 20-75 MB disk space per study (Bon Secours Health System, 2001).

2.9 Technical Network Basis

The widespread networking of today's computer systems - especially in HIS and teleradiology networks, means that they can share information with other computer systems. The connection that allows for such electronic communication is physically created by a cable or radio connection. In their entirety these connection elements are known as a network, whereby more web software also supports the transmission of information. An intranet is a network where the computer systems of a company or organization are connected with each other, while the Internet links intranet or individual computer systems worldwide (Balzert 2004).

Depending on the required information transfer speed the following transmission media are used for network construction

- Phone or Integrated Services Digital Network (ISDN) lines. The transmission speed of 14,400 bit/s to 64,000 bits/s corresponds to 1,800 characters per second or 8,000 characters per second
- Asymmetric Digital Subscriber Line (ADSL) - access over copper telephone or ISDN lines. Transfer speeds of between 768 kbps and 3072 kbps download and 128 Kbps and 384 Kbps for uploading
- Twisted-pair wire or coaxial cable at speeds between 10 Mbps and 100 Mbps
- Optical fiber or glass fiber. Transmission speed between 100 Mbps and 500 Mbps (Balzert 2004)

Information carried in networks is transmitted in packets, where the information is divided into one or more packets of fixed or variable length and transported over the network. Basic and broadband networks are distinguished depending on whether this transfer of packets takes place individually or in parallel. While information packets are transported in serial in baseband networks (Time Multiplexing or TDM for short), broadband networks can transmit several packages simultaneously. "This happens because the transmission medium, usually fiber optic, is loaded with several non-overlapping frequency bands. Each frequency band can transfer a packet independently of the other bands (frequency division multiplex)" (Balzert 2004).

The transmission protocols define the way in which a computer system is granted access to the transmission medium. Depending on the transmission medium and the transport protocol different network connection systems are required:

- An analog telephone network requires a modem (modulator / demodulator), which modulates the digital signals at the transmission point (bits), (converts into acoustic signals) and the receiver demodulates them (converts back to digital).
- A Digital Subscriber Line (DSL) modem is required for an ADSL connection.
- In the presence of a digital telephone network, up to eight different devices can connect to a basic line via an ISDN card, with up to two of them being used simultaneously. The narrowband ISDN allows data rates up to 64 Kbit/s, whereas the increasingly available broadband DSL system allows several multiples of that amount.
- If the computer system is to be connected to a LAN (local area network, LAN) a network card is required. LANs can connect computer systems in institutions or companies over short distances of a few meters to several kilometers, without using public connections (Balzert 2004).

A network can operate not only as an information network, but other resources, including printers and external storage, can be shared. In order to use these expensive peripherals effectively, they can be connected to the network as a server. A server is "a device that provides specified services for participants in the network upon request „ (Balzert 2004). They can be distinguished according to function, for example between print servers or where several archive servers act as a data warehouse.

In turn, the computer systems which require the services of the server are referred to as clients. A computer system can also simultaneously be a client for service, while also being available as a server for other services (Balzert 2004).

2.10 Teleradiology

Teleradiology is defined as being radiological examinations of a patient with spatial separation of the attending radiologists. In the narrow sense § 2 of the X-ray Ordinance defines teleradiology as the examination of a person with X-ray radiation under the responsibility of a physician pursuant to § 24, paragraph one, which is located not at the place of execution and is connected directly to the people at the site of the technical implementation by the use of electronic data transmission and telecommunications, in particular for the justification of indication and diagnosis (X-ray Ordinance 2002).

The definition can be applied to all modalities in which radiology images and multimedia information, such as video conferencing, ultrasonic sounds, language or control files are transmitted via a telecommunication device to a remote location.

Physicians describe the transmission of radiological images over a telecommunications device to a remote location as a teleradiology. This means that small hospitals that cannot afford their own radiologists do not have to go without the appropriate investigative methods. Competence Centers are also benefiting from the DICOM email standard and treatment planning can be carried out off-site. In addition to the data protection laws that prohibit the transmission of unencrypted patient data on public channels, the size of the radiological image data is of high importance for technical implementations (Engelmann et al. 2005).

The given basic requirements are the technical developments in telecommunications, the Internet and methods in medical image processing. The speed of data transmission lines such as ISDN (Integrated Services Digital Network) or DSL (Digital Subscriber Line) via telephone lines and high-speed network connections within a broadband-based LAN (Local Area Network) or WLAN (Wireless LAN) allow for multimedia data to be sent to any location with appropriate infrastructure in a reasonable time (see Section 2.9). The development of compression algorithms and communication standards have contributed to the facilitation and dissemination of teleradiology. Through the development of electronic patient records, besides the actual image data, information from the HIS and RIS can be transmitted in teleradiology networks.

Teleradiology allows radiological equipment to be shared. Even smaller hospitals or clinics are able to offer investment-intensive procedures such as computer tomography or magnetic resonance imaging without a radiologist being present on-site. Especially during night time, Sunday and public holiday service, this results in economic and procedural advantages. Wide area patient care, especially emergency care, can be improved, as with the application of teleradiology experts may be consulted over long distances. The advantage for patients is that in addition to diagnoses being safer they will be less likely to have to be transferred to another hospital for examination (Bode Meyer et al. 2002).

Another important and significant matter is the interdisciplinary cooperation between radiologists and clinicians. An example is the connection between the Department of Oncologic Diagnosis and Therapy of the German Cancer Research Center (DKFZ) in Heidelberg (Prof. Dr. G. van Kaick) and the internists and urologists of the Salem Hospital in Salem, Heidelberg (Prof. Dr. H. Seitz, Prof. Dr P. Müller, PD Dr. & Ikinger). The Salem Hospital sends patients for CT and MR examinations at the DKFZ and the images are transmitted via ISDN line to the teleradiology workstation in Salem (Engelmann et al. 2002). The doctors there can then access the data at any time. The radiologists at Salem Hospital regularly conduct telephone conferences with the clinicians at the DKFZ in order to discuss acute cases.

The government has set high hurdles for the use of teleradiology in the X-ray Ordinance law book, which often involves complicated and lengthy approval procedures. Another application of teleradiology is the tele-consultation, which has lower regulatory hurdles. In this form of teleradiology the radiologist responsible is always present where the tests are being carried out. In particularly hard to diagnose or cases where there is considerable doubt the radiologist responsible for the preparation of the findings can seek to obtain a further opinion, for example, from an expert at a university hospital.

The standardization of interfaces and data transfer and communication protocols, especially in tele-consultation is significant because the communication takes place with many different and previously established communication partners. In May 2004 the German X-ray Society (Deutsche Röntgengesellschaft) adopted a recommendation on the standardization of teleradiology in Germany. Following that recommendation the OpenPGP encrypted email DICOM standard should be used for image transmission in teleradiology (Engelmann et al. 2005, Mildenberger et al. 2005).

Teleradiological solutions have been proven to be productive through use for many years in various different locations. One example of a recent project which has used modern technologies and practices is the one implemented in 2004 known as the Rhein-Neckar-Triangle Teleradiology Project. The project was put into operation over a three-year construction phase. The data network is spread over twenty hospitals and medical practices in the area of Mannheim, Heidelberg, Karlsruhe and Ludwigshafen (Weisser 2005).

The special feature of the tele-radiology, and thus its special strength, is the communication of any number of agencies working in a network. An example is shown in Figure 9, the CHILI Teleradiology Network (Engelmann et al. 2002) with more than 60 installations in Germany and the USA. The CHILI teleradiology network also allows teleconferencing. The images are displayed simultaneously and can be edited directly by the visible mouse pointer of both communication partners.

illustration not visible in this excerpt

Figure 9: The CHILI teleradiology network (by Engelmann et al. 2002)

In the context of the impact of digital technology on the patient, these issues are advanced further (see Chapter 4.1.2).

2.11 System integration architectures and models

Information systems provide solutions and tools available in the form of architectures and models that support the integration process for the introduction of systems in new and existing information systems. The introduction of a new RIS for an existing HIS is a special case for this area of application.

2.11.1 Enterprise Application Integration (EAI)

The goal of EAI is the integration of application systems in an enterprise, for example, an existing HIS. According to Ackermann et al. the following four views for integration can be distinguished:

illustration not visible in this excerpt

Figure 10: The views of the EAI (according to ACKERMANN et al. 2006)

The specification for the integration of views takes place in a business-bus, which is attached via a software adapter with different instances within the views. This relationship is shown below.

illustration not visible in this excerpt

Figure 3: Enterprise Application Integration (according to WIKIPEDIA 2007)

Practical application of the EAI using selected HIS

In 1996, one example grew from the need for cooperation between the subsystems of the HIS of the Cologne University Hospital. With the help of a hospital communications server the HIS-internal functional coupling, based on the HL7 communication protocol, was created.

The following graphic illustrates the environment of the application scenario of the communication servers at the Cologne University Hospital (Heitmann et al. 1998):

illustration not visible in this excerpt

Figure 12: The hospital communication server of the University Hospital of Cologne

(By Heitmann et al. 1998, p. 105)

In the medical facilities of the Westfalian Wilhelms University in Münster, Germany, in the mid-1990s, there was a need for the hospital to conceptually renew its communication system in order to continue to provide high quality patient care and appropriate level of hospital management. The rapid availability of both departmental and cross- departmental information (patient data, findings from the functional areas, examination orders, etc.) was the main goals of these modernization measures. EAI was also used here (Osaka et al. 1998).

The underlying communication and interaction structure of the DATAGATE of the HIS of the medical facilities has a subsystem: "IDIK" for patient database management and billing, "OLIS" as a departmental system of the central laboratory, "ProTransmed" for transfusion medicine, RADOS for radiology and MacDoc for nuclear medicine .

illustration not visible in this excerpt

Figure 13: Interaction and communication structure based on Datagate

(By Osada et al. 1998, p.123)

Between the IDIK and OLIS systems there is a point-to-point connection. All other departmental systems draw all relevant HIS-wide data from these two departmental systems. The communication links function in Datagate as follows (Osada et al. 1998):

- All patient data is sent from IDIK to OMD
- Journal and archival reports are provided by OLIS to OMD
- Messages are transferred as files
- Asynchronous communication between the transmitter and receiver

Using EAI, the entirety of all messages and data are exchanged as digital files. These are easily executed by the linked web applications. According to Osada et al. the optimization of electronic reports transmission meant that most results were available three hours sooner and that doctors had to make significantly fewer telephone requests for reports (Osada et al. 1998).

2.11.2 Service oriented Architecture (SOA)

Where mainly various different Web services communicate with each other as part of an application, this is known as an SOA. The integration focus is directed to the services used. From a technological perspective, what is meant here by a ‚service‘, is a component that is equipped with a self-descriptive interface and which can be independently called up using standardized protocols (Ackermann et al. 2006).

According to Zielinski et al., services will be divided into categories for the representation of the system functionality and to provide the functionality required by the user (Zielinski et al. 2006). In addition, a SOA is seen as the sum of four different sub-components. These are as follows (Zielinski et al. 2006):

- A User Interface as the front end of the application
- The service itself
- A service repository as an archive of all possible services

- A service-bus, which acts as a channel for the service call

Each individual service is made up of agreements (contracts) with rules for service use, the implementation of the underlying business logic and the data basis and from ports.

A project modeled on the SOA system, developed by the University of Oslo, called "Service-Oriented Architecture and Web Services in Healthcare Application Production and Integration" (SerAPI) extends the SOA to include other concepts, with whose help the characteristics of applications and processes of the departmental IS can be adjusted individually to the local requirements of a HIS (Korpela 2004). The following illustration schematically shows the functionality of the SerAPI.



ISBN (eBook)
ISBN (Buch)
8.2 MB




Titel: The digitization of radiology and its impact on patients, processes and economics of modeling