Although the concept of picture archiving and communications systems (PACS) was developed in Europe during the latter part of the 1970s,1 no working system was completed at that time. The first PACS implementations took place in the United States in the early 1980s, eg at Pennsylvania University, UCLA, and Kansas City University. Some more or less successful PACS developments also took place in Europe in the 1980s, particularly in the Netherlands, Belgium, France, Austria, the United Kingdom, Italy, and Germany.2 Most systems could be characterized by their focus on a single department, such as radiology or nuclear medicine. European hospital-wide PACS with high visibility evolved in the early 1990s in London (Hammersmith Hospital) and Vienna (SMZO). These were followed during the latter part of the 1990s by approximately 10 to 20 PACS installations in each of the major industrialized countries of Europe.3
At this point, however, there are still far more working PACS installations in the United States than in Europe. There is no inherent technological reason why this should be the case. In the 1980s, there were already strong European companies that offered and could maintain PACS in hospitals. Indeed, European PACS vendors were more successful in the United States than in Europe. If not the lack of technology, then other reasons must account for this difference in the PACS installed base. An attempt is made in Table 1 to enumerate 10 possible factors that favor installation in the United States and hinder installation in Europe, with no claim of completeness or effort to prioritize any one of them (some of the reasons given in this table provide enough material for a monograph!).
There is a further overriding problem that radiology faces (not only in Europe but worldwide), which is that the new digital world shows, compared to the traditional analog world, fundamental differences in principle and degree in how our workplace is modeled. In the zeitgeist of a global information society, the traditional existence of radiology, based on the analog world of radiation physics and technology, film, light box, work flow, and patient is being fundamentally challenged. For more than 100 years, every patient could be treated on a very individual basis. Although the organizational radiological infrastructures were fairly static, the degree of freedom (DOF) in work flow to care for patients was very high indeed. In an analog world, no two situations or cases are the same and the differences cannot be exactly measured and controlled.
In a digital world, however, analog processes and attributes are reduced to ones and zeros. Dissimilar analog sources, when digitally converted, can be precisely defined by ones and zeros. In other words, if radiology becomes digital, everything becomes digitally representable, measurable, and controllable, with the question of who will be in control being left unanswered. This potential change in work environment may be an overriding reason why PACS, as part of a digital radiology, has been viewed with skepticism by some leading European radiologists.
When this analog world of radiology first met the digital world of informatics in the 1970s, only a few radiologists were prepared to look and move across their discipline’s boundaries. This was understandable, because the digital world in those times was very narrow, with little DOF. Informatic theories and tools emphasized functionality, von Neumann architectures, structured system design, CASE models, and efficient algorithms implemented in assembler or procedural high level languages. With improved accessibility of computers in the 1980s, information systems were still largely unusable for work flows in health care, particularly those that required a high degree of flexibility, exception handling, and human-friendly interfacing, needs that characterize health care processes such as radiological diagnosis.
To overcome some of these problems, a gentle way of introducing the digital world to the radiological community has been followed in Sweden with Pax Vobiscum, an intranet-based system for the modern radiology department.4 With the conviction that “digital radiology in its most extreme form, the filmless radiology department/hospital using a radiology information system (RIS)/PACS for acquiring, reviewing, distributing image information, has until recently not really been established as a routine phenomenon, but rather as a business for pioneers, visionaries, and crazy people in general,” emphasis has been given to providing PACS with added value to the users.4
The Pax Vobiscum project built one virtual filmless radiology department through a total digitalization and integration of five independent radiology departments within ?rebro County in Sweden. This virtual department performing approximately 180,000 procedures/year has been running routinely since 1997 and is at present one of the largest PACS installations in Europe, with 48 radiology rooms forming on e functional unit.
The added value refers to the process of changing the way in which information is distributed inside and from the radiology department. It includes, in addition to image transfer to other departments, information availability on patient management, price lists, education sessions, and basic patient information. After 3 years of operation, a summary of the project5 states: “The most important experience from this large project is the fact that the technical problems were not overwhelming but the human reactions and problems were larger than expected…. Over time people learned to benefit from the new technology and now, after three years in the digital reality, no one is willing to go back to analogue technology.”
|Figure 1. Ten possible favorable factors (for the United States) or hindering factors (for Europe) for the implementation of PACS.
Hammersmith Hospital and SMZO
A thorough examination of the impact of new digital technology before and after a PACS has been implemented is intended by many but completed by few. One such evaluation, however, took place for the hospital-wide PACS at Hammersmith Hospital in London. The main focus of this evaluation was hospital-wide availability of images and reports when required by different categories of clinicians, as well as the time devoted by junior staff to searches for images.6 The result of the survey showed “that PACS greatly reduced the perceived problem of image non-availability for both inpatients and out-patients, which benefited patients (significantly fewer examinations had to be repeated because the original image was lost) and also junior clinicians (they spent less time looking for the images).” Availability of radiological reports, however, was not as positively judged after PACS was operational. This extensive review took place between 1992 and 1996.
The other pioneering work in a hospital-wide PACS endeavor in Europe took place in Vienna in the hospital Socio-Medical Care Centre East (SMZO). Right from the beginning, PACS was integrated into the newly built hospital, which opened in 1992.7 With no possibility of a pre-PACS evaluation, some extended surveys in 1998, comparing the SMZO with other major hospitals in Vienna, showed that average length of stay and average cost of radiology services (excluding CT and MRI) were significantly lower at the SMZO (see Tables 2 and 3, page 34).
How these results are related to the impact of PACS as part of digital radiology is subject to speculation. After continuing technology upgrades of the SMZO PACS, internal reviews of the introduction of digital radiology3 observed many changes in daily routines, such as:
workloads are moved between employee groups
cycle times speed up significantly
employee qualifications increase
process descriptions are changing very fast; new capabilities allow radiologists new ways of diagnostic work.
Recent PACS developments
With new PACS architectures, technology, DICOM standard extensions, and, in particular, the Integrating the Healthcare Enterprise (IHE) initiative, objective external evaluations of PACS installations older than 2 years should themselves be carefully evaluated before being adopted as guidelines. For some of the reasons derived from Table 1, in-house evaluations should be handled with even greater caution. It is difficult to obtain an objective up-to-date evaluation of any of the 100 or more current PACS implementations in Europe in the spirit of the Hammersmith Hospital survey. There is little willingness to free the necessary funds from the PACS investment to engage an external agency to carry out a pre- and post-PACS evaluation.
Most modern PACS realizations are implemented in a step-by-step fashion. After successful conversion (judged internally) of one department from analog to digital, the next candidate department will be selected on a situational planning initiative. A typical example is the Innsbruck University Hospital.8 After PACS was realized for the traumatology department in 1997, soon other departments, such as surgery, followed. Bit by bit, also, teleradiology services were realized within the region of Innsbruck. With about 80% of radiology data now being handled with PACS, full PACS coverage for the whole hospital is planned for mid-2001.
Sometimes the process of converting to digital is accelerated and completed within 1 to 2 years. This is exemplified by recent PACS implementations in Paris at the Institute Gustave Roussi (IGR)3 and the City Hospital of Marzahn in Berlin (www. ukb.de). Very seldom is there the opportunity of a clean start as at the SMZO. The newly built Pompidou Hospital in Paris,3 which opened in July 2000, is an exception, but even here great difficulties were experienced in running a digital radiology department and a film-free hospital. For these and many other recently installed systems, an external evaluation is not readily available.
With the advent of improved multimedia user interfaces, local and global digital communication systems, informatic standards for middleware and applicationware, and models for information presentation and transactions, the gap between the analog and digital worlds is clearly closing. Standards or de facto standards allow informaticians to become more user-oriented and adapt their information systems to the needs of radiologists.
This development, combined with a PC-user-oriented generation of newly trained radiologists, and complemented by new cost models such as application service providers (ASPs), is giving rise to an increased interest in using PACS. With pay-per-use PACS applications, contrary to the situation in the past, expenditure for PACS technology and its maintenance as well as its impact on departmental fixed costs are becoming less of a primary concern. Rather, there is increased awareness of radiological work flow and practice patterns to improve services to other departments, hospitals, clinics, and referring physicians. This is the birth of regional PACS.
Taking the first 20 years of European and US PACS experience into account, there is also in Europe a move toward more strategic planning, rather than following a situational solution-seeking approach. In central Europe, this is exemplified by a reference project in Saxony in Germany (Saxtelemed).9 This regional PACS and telemedicine project is aimed at improving health care in Saxony and providing guidelines to other hospitals in this state (and perhaps other federal states in Germany and elsewhere). Seven major hospitals and many other connecting health care institutions and physician practices are being heaved into the digital age. Focusing with the first investment plan of DM 28 million between 1999 and 2001 on PACS and telemedicine, a very rigorous request for proposal (RFP) procedure for vendor selection has been developed.10 It includes extensive workstation evaluations for radiology and clinical applications. Some of the RFP methodologies developed are already being used for PACS planning for other hospitals, including Zurich University Hospital. External pre- and post-PACS evaluations are in progress based on the Marburg Model. Other strategic features of the Saxtelemed project include the introduction of the health professional card (HPC) and the first studies on the feasibility of a (virtual) electronic patient record (EPR).
|Figure 2. Comparison of average length of stay and average cost per inpatient at major Vienna hospitals. Source: WrKAV-Leistungsbericht 1998. 1 US$ = 15.53 Austrian Schillings.
The Saxtelemed project, however, is not unique. Similar endeavors for wide area or regional PACS can be found, for example in Southern Europe.11 In Siena, Italy, a RIS/PACS project implementation is being carried out at a provincial level, including four heterogeneous hospitals and two health care centers, representing approximately 290,000 radiological procedures per year. The implementation of RIS/PACS is seen as a highly complex project, which introduces a cultural change in the radiology department’s organization and management. It is also stated that “work flow model comparisons (analog and digital) require a global radiology reengineering in terms of redrawing the distribution of physical spaces, redefining the human resources allocation, redefining the data, information and image flows, and replanning the staff and patient flows.”11? Another project in Hordaland County in Norway integrates eight hospitals to facilitate transmission of images and data between their radiological departments.13 The county is expected to carry out more than 350,000 procedures per year for improved patient care in this “seamlessly integrated system.”
These examples make it very clear that PACS can no longer be seen in isolation from the rest of information technology (IT) in health care. Images are only part of the story, with the EPR increasingly becoming the center of attention.
|Figure 3. Comparison of average cost of radiology services (excluding CT, MRI) at major Vienna hospitals. Source: Bundesmin. fur Arbeit, gesundheit u. Soziales 1998. 1 US$ = 15.53 Austrian Schillings.
For health care environments that encourage or allow a strategic approach, the widespread introduction of an electronic health care record (eg, National Health Service in the United Kingdom14) is seen as a prerequisite for modern health care. This will eventually promote PACS in order to incorporate images into the EPR. From an IT perspective, however, this is only meaningful if it is realized in the context of an IHE-like initiative. Fortunately, the IHE technical framework initiated by the RSNA/Healthcare Information Management Systems Society consortium is given considerable visibility by European associations. If some other European activities and results, for example the Electronic Health Record Architecture standard proposals such as CEN TC 251 prENV 13606 (European Committee for Standards), are linked to the IHE, Europe may even lead the way (at least conceptionally) to integrate PACS into health care.
Europe is certainly endeavoring to be at the forefront in mobile medicine, by extending the functionality of PACS to mobile teleradiology. Building on experiences gained with the CHILI system (a software architecture that provides a PACS and teleradiology infrastructure15), a project on mobile teleradiology called MTM (Multimedia Terminal Mobile) is being funded by the European Union (EU).16 The CHILI radiology network has more than 50 medical installations in Germany and seven installations in the United States. Extending the CHILI concept, mobile teleradiology software will use personal digital assistants (PDAs), pen-based computers or Webpads for wireless access, based on Universal Mobile Communication System, to the medical images stored in the archive of a PACS or DICOM compatible workstation. In a survey carried out mostly in radiological departments in Spain and Germany, the application scenarios envisaged are:
Neurosurgical emergency consultations
Increased access to senior staff
Bedside access to patient data
Radiologist in contact with modality
Access to functional MR specialist
The next software version will be ready for evaluations in daily routine scenarios in June 2001 and will be presented at the CARS 2001 congress in Berlin.
Europe is also progressing very well as far as digital imaging for image guided therapy is concerned. With Food and Drug Administration legal restrictions that are tougher than European CE regulations, many US companies are looking for Europe to gain clinical experience with image-guided therapy systems. To fulfill the surgeon’s need for real-time images requires a type of service for which PACS can provide a suitable infrastructure with the EPR as a key multimedia information access technology. This means that increasingly PACS becomes the business of everybody involved in health care, including the patient.
Who uses and controls PACS for the many infrastructure services that it will finally have to fulfill is also difficult to predict. Health care information systems have traditionally been under the control of IT staffs and it is likely that PACS will become part of them. The more easily these systems can be used, the more complex is the middleware on which the applications and the human/computer interfaces are built. Specification, selection, and management of the complexity of these ones and zeros are? IT problems.
Traditional European culture, often based on protectionism, will have to give rise to more openness in health care. New service and radiology information centers may have to be introduced. With the European Commission and EAR sponsored Eurorad project, first steps are being taken in this direction. Eurorad is accessible on the net (www.eurorad.org) and provides a broad spectrum of radiological database/electronic teaching files covering all radiological subspecialties. It constitutes a suitable base for future computer-aided diagnosis and knowledge management systems.
Heinz U. Lemke, PhD, is head of the Department of Computer Graphics and Computer Assisted Medicine, Technical University of Berlin.
- Horii SC, Blaine GJ, eds. Medical Imaging 1997: PACS Design and Evaluation: Engineering and Clinical Issues. Proceedings of SPIE. 1997;3035.
- Lemke HU, et al. Medical Imaging 2000: Early Developments of PACS in Europe. Proceedings of SPIE. 2000:3980.
- Caramella D. EuroPACS Web site.
- Andersson T. Introduction of PACS from a clinical viewpoint. In: Gell G, Holzinger A, Wiltgen M, eds. Proceedings of the 18th International Conference EuroPACS 2000: PACS to Internet/Intranet, Information-Systems, Multimedia and Telemedicine; September 21-23, 2000; Graz, Austria.
- Andersson T. Intranet as a quality tool for the modern radiology department? the Pax Vobiscum Information System. In: Gell G, Holzinger A, Wiltgen M, eds. Proceedings of the 18th International Conference EuroPACS 2000: PACS to Internet/Intranet, Information-Systems, Multimedia and Telemedicine; September 21-23, 2000; Graz, Austria.
- Bryan S, et al. The benefits of hospital-wide picture archiving and communication systems: a survey of clinical users of radiology services. BMJ. 1999;72.
- Hruby W, Maltsidis A. A view to the past of the future?a decade of digital (r)evolution at the Danube hospital. In: Hruby W, ed. Digital (R)evolution in Radiology. Vienna: Springer Publishers; 2000.
- Vogl R. Medical Imaging 2001: Going All Digital in a University Hospital: A Unified Large Scale PACS for Multiple Departments and Hospitals. Proceedings of SPIE. 2001;3290.
- Wedekind R, R?sler R. The SaxTeleMed model programme?requirements and goals from the standpoint of the Ministry of Health of Saxony. In: Lemke HU, Vannier MW, Inamura K, Farman AG, Doi K, eds. CARS 2001 Proceedings of the 15th International Symposium and Exhibition, Computer Assisted Radiology and Surgery. Excerpta Medica, International Congress Series 1230. Amsterdam: Elsevier Publishers; 2001.
- Lemke H, et al. Request for Proposal for PACS and evaluation of tenders. In: Hruby W, ed. Digital (R)evolution in Radiology. Vienna: Springer Publishers; 2000.
- Caliani M, et al. First implementation of a RIS/PACS project of provincial level in Italy: planning methodology and expected results. In: Lemke H, Vannier MW, Inamura K, Farman AG, Doi K, eds. CARS 2001 Proceedings of the 15th International Symposium and Exhibition, Computer Assisted Radiology and Surgery. Excerpta Medica, International Congress Series 1230. Amsterdam: Elsevier Publishers; 2001.
- Stefani P, et al. Filmless and paperless management of a radiology department: a new work process and control system. In: Lemke H, Vannier MW, Inamura K, Farman AG, Doi K, eds. CARS 2001 Proceedings of the 15th International Symposium and Exhibition, Computer Assisted Radiology and Surgery. Excerpta Medica, International Congress Series 1230. Amsterdam: Elsevier Publishers; 2001.
- 8 Million Euros PACS/RIS deal. European Hospital. 2000;9(5).
- An Information Strategy for Modern NHS 1998 – 2005. Available at: www. nhsia.nhs.uk.
- Engelmann U, et al. Evaluation of the CHILI teleradiology network after 3 years of clinical routine. In: Gell G, Holzinger A, Wiltgen M, eds. Proceedings of the 18th International Conference EuroPACS 2000: PACS to Internet/Intranet, Information-Systems, Multimedia and Telemedicine; September 21-23, 2000; Graz, Austria.
- Engelmann U, et al. Mobile teleradiology: all images everywhere. In: Lemke HU, Vannier MW, Inamura K, Farman AG, Doi K, eds. CARS 2001 Proceedings of the 15th International Symposium and Exhibition, Computer Assisted Radiology and Surgery. Excerpta Medica, International Congress Series 1230. Amsterdam: Elsevier Publishers; 2001.
- Stefani P, et al. Implementation of PACS and informatics in Department of Radiology of Siena University: a cost/benefit analysis. In: Lemke HU, Vannier MW, Inamura K, Farman AG, Doi K, eds. CARS 2001 Proceedings of the 15th International Symposium and Exhibition, Computer Assisted Radiology and Surgery. Excerpta Medica, International Congress Series 1230. Amsterdam: Elsevier Publishers; 2001.