|
|
|
 |
|
 |
 |
| Davide Caramella is Associate
Professor of Radiology in the
Department of Oncology,
Transplants and New Technologies
in Medicine of the University of Pisa.
He is Past President of EuroPACS
(http://www.europacs.org),
Chairman of the Information
Technology Committee of the
European Association of Radiology
(http://www.ear-online.org),
Scientific Director of EURORAD
(http://www.eurorad.org) and a
Member of the Editorial Scientific
Board of European Radiology.
Dr Caramella has published over
100 papers and chapters on
magnetic resonance imaging,
teleradiology and PAC systems.
He is the co-editor of four books on
computer applications in radiology,
advanced image processing and
radiological resources
on the Internet. |
The implications of digital imaging for
clinical practice
Davide Caramella
Department of Oncology, Transplants and New
Technologies in Medicine, University of Pisa, Italy
Address for correspondence:
Professor Davide Caramella
Associate Professor of Radiology
Department of Oncology, Transplants and New
Technologies in Medicine, University of Pisa
Via Roma, 67, 1-56100 Pisa, Italy
Tel: +39 050 992 509 Fax: +39 050 551 461
Email: caramella@do.med.unipi.it
Introduction
The introduction of digital imaging over 30 years ago
has dramatically broadened the clinical applications of
radiology. Not only have new radiological modalities
emerged, making it possible to study finer anatomical
details, elucidating pathology as well as function, but
the workflow in radiology departments has also
been revolutionized.
The Picture Archiving Communication System (PACS)
concept was first proposed in 1982. The expectation
was that it would solve most of the problems of
conventional radiology by allowing efficient image
management and thus achieving organizational and
economic advantages, ultimately improving patient
care. In reality, things were not so simple, mainly
because the technology of the time was too immature
to allow the implementation of filmless operations
within hospitals.
PACS evolution
The paradox of first-generation PACS was that these
systems were planned by a technologically motivated
computer scientist at a time when information
technology resources were still painfully inadequate for
meeting the stringent requirements of the practice of
radiology. Other reasons contributing to the failure of
first-generation PACS were the legal framework in
many countries (which were unprepared to cope with
the novelties of the digital revolution in medicine),
widespread customer dissatisfaction with the
rudimentary systems that were initially introduced to
the market and - last but not least - the high cost of
the systems.
The evaluation of health care costs is not an easy
subject and there are several areas of controversy.
These include the determination of benefits (both
payable and intangible) and the inclusion of indirect
(or ‘hidden’) costs and benefits.When evaluating PACS, the most common pitfall is to calculate costs and
estimate savings taking into account only the radiology
department. This narrow evaluation will reveal only that
a changeover to digital systems entails additional costs,
with very few savings to be expected (savings in films
alone are of course insufficient to justify the massive
investment needed to implement PACS from scratch).
However, such an evaluation neglects the substantial
savings in terms of the increased overall efficiency of
the health care institution that can be achieved thanks
to the integration of the digital management of all
clinical data (including images).
Moreover, there is a rather intriguing and peculiar
economic situation in medicine. In all other economic
sectors, such as industry, banking or insurance,
investments in information technology lead to an
increase in productivity and ultimately to greater
profits. In medicine, on the other hand, the investment
in information technology seems only to increase the
costs, since the revenues of hospitals are largely
independent of their productivity.
Many things have changed since the early years of
PACS. Today powerful personal computers have replaced
the more costly UNIX machines and provide adequate
computing platforms for almost all medical applications.
Storage media have evolved, giving us the opportunity
to archive a huge amount of data at only a fraction of
the cost of a few years ago. Network technology has
been revolutionized, with a marked increase in the
available bandwidth and a progressive convergence of
local area networks and wide area networks, as well as
the introduction of wireless and mobile
telecommunication. Standards such as Digital Imaging
and Communication in Medicine (DICOM) and
Integrating the Healthcare Enterprise (IHE) (Figure 1)
have finally made it possible to achieve multi-vendor
inter-operability across hospital information systems.
Both national and European laws have evolved taking
account of new issues such as digital archiving, the existence of databases containing confidential
information and the importance of privacy and security
in the digital environment.
All these important changes were prerequisites for the
development of the second-generation PACS that have
now proven successful in the clinical domain. Today
technology is no longer a limiting factor, as highperformance
networks are available, standards are
universally accepted and national and international
laws have been modified to meet the needs arising
from the digital revolution.
Radiological reporting
The production of the report is one of the main goals
of the entire radiological process and PACS has a
relevant (and rapidly evolving) role in this area.
The introduction of soft-copy reporting and voice
recognition has changed the way radiology is practised.
However, there are still questions that need to be fully
addressed, such as the diagnostic performance of the
radiologist (i.e. reduction of mistakes) and ergonomics
(i.e. workstation fatigue), among others.
Without soft-copy reporting, it would not be possible
to read effectively the large volumes of data produced
by up-to-date acquisition modalities, nor would we
have experienced the seamless integration of advanced
image processing, 3-4-5-D image analysis, multimodality
image fusion, treatment planning and computer-aided
diagnosis into the radiological workflow (Figure 2).
Finally, reporting at the PACS workstation has made it
possible to foresee the systematic use of structured
reports, with the potential for reducing the indeterminate
nature of many radiological reports, making the
information included in our reports ‘databasable’ items
that must be entered unambiguously in order to
facilitate all kinds of ‘data mining’ at a later date.
 |
| Figure 2. Image processing in vascular imaging and interventions. |
The impact of these structural changes on the outcome
of the radiological process has not yet been fully
understood. There is a risk that radiologists will be but
passive acceptors of the changes being imposed by
technology, rather than being active in planning for
changes and steering their implementation. The
structured report will change the way radiologists
interact with referring colleagues, streamlining the
information flow. PACS has already changed this
complex relationship - for example, with the distribution
of electronic images within hospitals it is clear that the
pattern of utilisation of images by clinicians has
changed but the details of these changes are still to
be fully elucidated.
Radiologists may claim that they provide added value
to clinicians by expediting the distribution of images
and reports throughout the hospital. But how is this
quantified? Have outcome studies been published that
justify this assumption? Has the potential for more
efficient hospital-wide PACS-mediated informationsharing
to reduce medical errors and achieve better
patient care been realized?
The distribution of radiological images is no longer
confined within the hospital, since in many instances
regional PACS are emerging as the best solution for
a rapidly consolidating health care sector. This trend
will make the term ‘teleradiology’ obsolete, since
teleradiology is progressively becoming just another
function of PACS. From a technical point of view, image
distribution has been implemented using the web
architectural model or the integration of off-line
devices, such as CDs, which are often involved in
communication with general practitioners. Moreover,
many innovative products from the consumer market
(e.g. latest-generation mobile phones, tablet PCs, hard
drive-based devices) may be tested and adopted for
improving the distribution of radiological data.
 |
| Figure 3. Outsourcing to an Application Service Provider (ASP) |
One interesting business model developed in the service
industry (e.g. travel, insurance and banking) that
is usefully being adopted by radiologists is ASP
(Application Service Provider). An ASP operates software
at its data centre that customers access on-line under a
service contract. In radiology this can easily be applied
to the archive, which can be ‘outsourced’ to an ASP
vendor (Figure 3).
Future of PACS
Radiologists are medical professionals who work with
images. However, many other medical disciplines are
based on imaging or require doctors to conduct imageintensive
tasks (surgery, for example). Most PACS
include nuclear medicine and radiotherapy, but a great
deal of investigational effort will be necessary to define
and then meet the specific requirements for integrating
pathological images (interactive display of the
microscopic fields), endoscopic images (video
sequences) and dermatological images into the PACS
environment. And still more research will be needed
to specify and implement so-called ‘surgical PACS’.
All hospitals with PACS have experienced their ability
to improve the quality of teaching, both in terms of
continuing medical education (e.g. clinico-radiological
conferences) and in the specific area of radiological
training. The availability of large numbers of images and amounts of clinical data allows ready access to
pathological examples, facilitates the construction of
multimedia teaching files and prepares physicians to
use the powerful resources of e-learning.
We have recently witnessed an impressive increase in
the radiological resources available on the Internet,
many of which are maintained by universities or by
ad-hoc organizations that have gained great visibility
on the Internet (examples include AuntMinnie.com,
Medcyclopedia™ and CTisus.com).
There have been concerns about the diffusion of
medical information on the Internet, mainly due to
the potential presence of unchecked errors and to the
possible misuse by patients of complex medical data.
One interesting approach comes from EURORAD, the
e-learning initiative of the European Association of
Radiology. It has made over 1,500 peer-reviewed
teaching files (many of them translated into French and
Spanish) freely accessible at www.eurorad.org
(Figure 4).
The local radiological archive has ceased to be solely a
store for previous examinations kept for legal reasons
but has become an active repository of our professional knowledge that is updated at every encounter with the
pathologies that we correctly diagnose. In the book
PACS and Imaging Informatics, H K Huang has recently
introduced the intriguing concept of a medical imaging
informatics infrastructure that is designed to take
advantage of existing PACS resources, and their image
and related data, for large-scale horizontal and
longitudinal clinical service, research and education
applications that, due to insufficient data, were not
previously possible.
This is the future of PACS: one in which traditionally
separate decision-support modules will be fully
integrated, as well as new developments in the areas
of improved reporting strategies, optimized distribution
of radiological information to referring doctors and
knowledge management applications ranging from
e-learning to computer-aided diagnosis.
|
|
|
