E u r o p e a n C o m m i s s i o n

KI-NA-19-957-EN-C
Community Research

Optimisation of
Protection in
the Medical Uses
of Radiation

OFFICE FOR OFFICIAL PUBLICATIONS

OF THE EUROPEAN COMMUNITIES EURATOM
L-2985 Luxembourg

EUR 19793
Medecine.EN (Flash) 23/10/02 16:34 Page 2

Editors

Dr. Ernst-H. Schulte, Dr. G. Neale Kelly, Ms Karin Coiffard

Contact person: E.H. Schulte, EC DG RTD J04

Address: Office MO 75 5/4, B-1049 Bruxelles/Brussels
E-mail: ernst-hermann.schulte@cec.eu.int

Legal notice

Neither the European Commission nor any person acting on behalf

of the Commission is responsible for the use which might be made
of the following information.

A great deal of additional information on the European Union is

available on the Internet. It can be accessed through the Europa server
(http://europa.eu.int).

Luxembourg: Office for Official Publications of the European

Communities, 2002

ISBN 92-894-3838-X

© European Communities, 2002

Reproduction is authorised provided the source is acknowledged.

Printed in Belgium
Medecine.EN (Flash) 23/10/02 16:34 Page 3

4 PREFACE

5 INTRODUCTION

Contents
12 J. F. MALONE Digital imaging: measures for optimising radiological information content
and dose

14 K. FAULKNER Measures for optimising radiological information and dose in digital imaging
and interventional radiology

16 J. GELEIJNS Computed tomography – techniques, image quality and patient dose

18 K. JESSEN Quality criteria for computed tomography

20 B.M. MOORES Unification of physical and clinical requirements for medical x-ray imaging

22 R. ORAVA New detector technologies in x-ray diagnosis and impact on radiation

protection strategies

24 R. JACOBS Maxillary imaging network on optimisation strategies and quality assurance

of radiographs in Europe

26 B. WALL Radiation protection of the patient in paediatric radiology

28 C. MACCIA Optimisation of radiation protection of the patient in paediatric radiology

30 H. ZOETELIEF Criteria and protocols for central laboratories providing dosimetric services
for quality control in diagnostic radiology
Medecine.EN (Flash) 23/10/02 16:34 Page 4

Preface

This brochure has been prepared to disseminate to a broad non-

specialist audience the objectives and main results of research
supported by the European Commission on radiation protection in
the medical applications of radiation. The brochure focuses on
the research carried out under the 4th and 5th Framework Programmes
but also addresses earlier research. This wider perspective should
enable the reader to gain a better understanding of the origins of,
and need for, research in this area. Some consideration is also given
to future research needs.

The brochure will appeal to those with an interest in, or responsibility

for, radiation protection in the medical applications of radiation. The
projects described encompass the ‘state of the art’ in this area together
with progress that can be expected within the next few years. Particular
attention is given to the development of new and improved tools
and methods for a better protection of the patient against undue
high radiation exposure during treatment or routine examinations.

The brochure is divided into two distinct parts. The first provides a
summary of Commission supported research over the past decade,
with particular attention given to the objectives and achievements
(or expected achievements) of the 4th and 5th Framework Programmes.
Remaining problems or challenges are identified and how these
may be addressed by future research. The second part contains short
summaries of the objectives and achievements (or expected achieve-
ments) of each of the supported projects.

Further information on each of the projects described in this report

can be obtained from the project consortia and contact points are
indicated for this purpose.

4
Medecine.EN (Flash) 23/10/02 16:34 Page 5
Summary
This brochure summarises the research on radiation protection
in the medical applications of radiation supported by the
European Commission under the 4th and 5th Framework Programmes.
The brochure concentrates on current research but places this
also in the context of earlier research and general developments
in medical applications of radiation. This wider perspective should
enable the reader to appreciate the origins of, and need for,
research in this area, and to put the results into context.
The brochure is divided into two parts. The first provides a back-
ground to the uses of radiation in medicine, especially the changes
over the last ten to twenty years and those envisaged for the future.
Against this background the reasons for the direction of the
Commission supported research over the past decade are clarified
and the research carried out is summarised. In broad terms much
of the research is concerned with underpinning the development
and implementation of the Council’s Directive on Medical Exposures
(97/43/EURATOM) issued in 1997. This Directive repeals and replaces
the earlier Directive (84/466/EURATOM). Some ideas for changes
in direction and emphasis for future research are given linked to
the changes in application foreseen. The second part contains short
summaries of each of the supported projects of the 4th and 5th
Framework programmes.
The participation of many of Europe’s main medical research
organisations has been instrumental in the success of this research
programme. Significant cost savings have been achieved through
the sharing of understanding and the spreading of measurements
throughout representative facilities in many countries. Participation
has served to consolidate and improve the focus of national
research programmes and the multinational interactions have
contributed to the widespread adoption of results and recom-
mendations reflecting best practice in a European context.
The main problem faced in the correct application of radiation
protection principles to radiology is to balance the need for diag-
nostically useful images against the need to keep radiation expo-
sures as low as reasonably achievable (the ALARA principle). It is
towards the solution of this basic problem that most of the research
described in this brochure has been targeted.
Commission supported research has led to major advances covering
both aspects of the basic problem in diagnostic radiology referred
to above. Years of direct clinical experience enable practitioners
to distinguish good images from those less useful and to decide
when an image is good enough for the purpose. It has been
necessary to distil the results of this clinical knowledge so that mea-
surable and reliable indicators can be developed for research and
practical application. At the same time efforts have been needed
to decide what measures of the radiation exposure are the best
determinants, especially for paediatric radiology, and to decide on
appropriate values that can be regarded as optimised for specific
examination types. A very important and clinically useful initial out-
come from the earlier work was the set of four European Guidelines
on Quality Criteria and Dosimetry published from 1996 to 1999.
Much has been achieved to promote the maximum benefit to
patients from medical applications of radiation, especially in diag-
nostic radiology. More research is needed to keep up to date with
rapidly evolving new diagnostic techniques and applications. Most
of the research carried out to date has concentrated on diagnostic
uses of radiation rather than therapeutic. In future there may
be merit in broadening the research scope to include radiation
protection in therapy, especially accident prevention.
PART 1
1. Introduction
The earliest and still the most common application of radiation is
in medicine. Radiographic images were in widespread use by the
turn of the nineteenth century, only 5 years after the discovery of
x-rays by Röntgen. This diagnostic use developed and expanded
during the whole of the twentieth century and was supplemented
by therapeutic use so that medical uses now contribute 14% of the
total exposure of people to radiation worldwide and over 95% of
the exposure to man-made radiation. About 100 million medical x-rays
are carried out each year in countries such as France and Germany.
There is no doubt of the enormous benefit from x-rays and newer
techniques in the diagnosis and treatment of accidents and diseases
of many types. Nonetheless the basic principles of radiation pro-
tection are that all exposures to radiation must be justified in the
sense that they should do more good than harm, and that the
exposure from all justified exposures should be as low as reason-
ably achievable. These basic principles apply to medical applications
as much as to any other applications but their application in clini-
cal practice is not straightforward. It was to develop methods by
which the principles can be implemented in all areas of medical use
that the Commission commenced research in this field.
2. Medical uses of radiation
The most common diagnostic technique is still the use of x-rays
to produce an image on photographic film. One of the earliest dis-
coveries that significantly reduced radiation doses was that a rare
earth image intensifying screen could be placed before the film cas-
sette. This technique was introduced in the 1970’s and is now stan-
dard practice. Dynamic images are needed for some applications
such as interventional radiology. These also utilise an image inten-
sifying fluoroscopic screen that is viewed in real time by the med-
ical practitioner, but in some cases cine film or more recently digital
recording techniques are used. The visibility of particular tissues can
be improved by introduction of contrast media such as barium for
the gastrointestinal tract or iodine for the urinary system.
The most far-reaching change in diagnostic imaging was to intro-
duce computer analysis and processing of digital images. This tech-
nique, known as Computed Tomography (CT), was first brought
into clinical practice in the1970s. It is able to provide very high qual-
ity images of isolated slices of the patient using a thin rotating beam
of x-rays, but the dose to the patient from a CT scan is much higher
than from a conventional x-ray. Because of the excellent images pro-
duced, CT has rapidly become an important imaging modality. In
European countries it now accounts for around 5% of all x-ray
examinations and 30-40% of the collective dose. The high doses
meant that a major component of the research programme was
directed to the possibilities for dose reduction or control in CT.
Developments in CT continue to be made with the introduction of
helical scanning and CT real-time fluoroscopy.
5
Medecine.EN (Flash) 23/10/02 16:34 Page 6

Box 1 : Definition of radiation dose

Radiation dose: Radiation deposits energy in tissue and causes damage. The harm from exposure of people to radiation can be
estimated from the radiation dose, which is a measure of the energy deposition.

In medical procedures it is normally the dose to a particular organ or tissue that is of concern. This is strictly known as the “equivalent
dose”. The unit of equivalent dose is the milliGray (mGy).

The “effective dose” takes into account the type of radiation, exposures of all or part of the body and the different radiosensitivities
of organs or tissues. The unit of dose used is the milliSievert (mSv).

Both of these will generally be abbreviated to “dose” in this brochure. Which quantity is meant can be seen from the context.

For some comparative purposes it is useful to add all the doses to a population, or all the doses from a particular examination.
In this case the result is referred to as the “collective dose” and the unit is the manSievert.

A different technique used in diagnosis relies on the administration
of radiopharmaceuticals to patients. This is generally referred to as
Nuclear Medicine and the measurements are made while the
material is in the body using an imaging device commonly called a
Gamma Camera. This can follow the movement of the radioactive
material through organs and tissues as a sort of biological tracer.
The behaviour in the body is a characteristic of the pharmaceutical
used with the incorporated radionuclide acting as the means of
detection and imaging. As new radiopharmaceuticals are being
introduced into clinical practice all the time, much research is directed
to optimising activity levels and estimating the dose from them.
The other quite different application of radiation is in cancer treat-
ment with radiotherapy. In this the procedure is to deliver extremely
high doses of radiation to the tumour with the objective of killing
it, while sparing the surrounding healthy tissues as far as possible.
The dose to the tumour has to be just enough to kill it but no more
or the damage to the rest of the body may be unsupportable, so the
objective is precision of dose delivery both in magnitude and loca-
tion. Radiation protection in radiotherapy calls for minimisation of
the dose to surrounding tissues and, on a rather different concept,
for the prevention of accidents. An accident in radiotherapy is an
event, such as miscalibration of the beam, leading to serious depar-
tures from the prescribed dose which could in turn lead to the death
of the patient. Radiotherapy is principally carried out using exter-
nal radiation beams from x-ray sets or sealed sources. There is
another modality known as brachytherapy in which radiation
sources are positioned on or introduced into the body so as to
deliver very high localised doses to the tumour. Therapy can also be
carried out using radionuclides introduced into the body as in nuclear
medicine. The current practice of using generalised treatment pat-
terns, which ignores individual radiosensitivity, may result in too much
or too little exposure of both the tumour and surrounding tissue with
implications for treatment success and for radiation protection. To
take this into account in routine operations may require the appli-
cation of underlying radiobiological research in a clinical context.
3. Development of the research programme up to 1994
The Commission first funded research projects in the field of radi-
ation protection in medicine in the late 1970’s. At that time there
was a lack of information on the doses to patients even from com-
mon diagnostic examinations. The initial research projects there-
fore concentrated on large-scale surveys of doses to patients from
a range of examinations, complemented by the development of
appropriate measurement techniques and mathematical model-
ling for organ dose assessments. At that time the staff using radi-
ation were more concerned with obtaining the diagnostic
information and the best image quality than with the level of dose
to the patient. One of the striking results from the surveys was that
the doses measured for similar examinations varied by up to a fac-
tor of a hundred, indicating considerable scope for dose reduction
without loss of diagnostic quality.
In 1984, a Council Directive [1] was issued laying down basic mea-
sures for the radiation protection of persons undergoing medical
examination or treatment. This Directive gave further impetus to
research as guidance was needed for its implementation.
The underlying objective of the research priorities chosen during
the 1980’s was to increase awareness of the need to take patient
dose into account and to establish links between the clinical and
technical requirements of the medical staff and the application of
radiation protection principles. The range of topics studied over this
period consequently included general aspects of quality assur-
ance and dose reduction measures, together with further dose sur-
veys and risk assessments.
In parallel with the research but using the initial results from it
initial attempts were made to define quality criteria for adult
patients for some common examinations. This resulted in the pub-
lication as a working document in 1990 of quality criteria for image
quality, imaging performance including patient dose and radi-
ographic technique.

6
Medecine.EN (Flash) 23/10/02 16:34 Page 7

PART 1

The emphasis on radiation protection in medicine was reinforced
by the publication in 1990 of new Recommendations of the
International Commission on Radiological Protection [2]. These
recommendations, which are still in force, clarified the concept of
optimisation of protection or ALARA and in particular intro-
duced the concept of reference levels (called investigation levels)
in diagnostic radiology that could serve as indicators of good
practice, or conversely of situations where improvements could
be made. This publication significantly influenced the direction of
the research programme during the 1990’s and indeed the prin-
cipal focus of the research programme for the period 1990-1994
was on optimisation of radiation protection in medicine.

Box 2 : The recommendations of the ICRP

The 1990 Recommendations of the ICRP: The principles for radiation protection in medicine [2].

The justification of a practice: No practice involving exposures to radiation should be adopted unless it produces sufficient benefit
to the exposed individuals or to society to offset the radiation detriment it produces. Most of the benefits and detriments accrue to
the individual patients and each procedure should be justified.

The optimisation of protection: The magnitude of the individual doses, the number of people exposed, and the likelihood of incur-
ring exposures where these are not certain to be received should all be kept as low as reasonably achievable, economic and social
factors being taken into account. Consideration should be given to the use of investigation levels for application in some common
diagnostic procedures.

Research projects under this programme during the period 1990-
1994 were grouped under two main headings:
Quality Assurance and Dose Reduction which included projects on:
◗ Optimisation of image quality and reduction of patient exposure
in medical imaging
◗ Quality criteria and dose reduction in CT and paediatric radiol-
ogy
◗ Digital medical imaging; optimisation of the dose for the exam-
ination, and
◗ Reduction of dose in x-ray diagnostics by the choice of the opti-
mal screen film systems.
Dose and Risk Assessment which included projects on:
◗ Medical dose assessment and evaluation of risk
◗ Patient dose from new pharmaceuticals
of other specific examinations additional work was carried out for
mammography, paediatric radiology and for the developing modal-
ity of CT.
Recognising that for the research to find practical application it was
necessary for it to be disseminated to and accepted by clinical med-
ical practitioners, a larger number of such scientists were involved
in the projects and workshops were held to spread knowledge of
the results being achieved. However it was recognised at the incep-
tion of the 4th Framework Programme that there needed to be
more consideration of the interaction between clinical needs and
dose reduction.
4. Research in the 4th and 5th Framework Programmes
The 4th and 5th Framework Programmes (FP4 and FP5) cover the peri-
ods 1994-1998 and 1998-2002 respectively. The primary objectives
of the two programmes followed on from the needs identified in
the previous section. Under a general title for FP4 of Optimisation
strategies for radiation protection of the patient in diagnostic radiology,
which made clear the focus of the research, one of the main objec-
tives was to:

◗ Evaluation of dose and risk due to interventional radiology
techniques, and
◗ A comparative investigation of diagnosis related doses in some
European hospitals.
Establish relationships between the quality of the radi-
ological information content with the image produc-
ing procedures, equipment functioning and
mechanisms of image analysis or interpretation – and
with the dose to the patient.

Building on the earlier work to develop quality criteria for adult
patients and the new concepts put forward by the ICRP much of
the work during this period concentrated on the creation of a pre-
liminary set of “dose reference levels” for major examination types.
To do so needed the results of dose surveys and the development
of technical means for dose reduction. In view of the importance
This very important objective made clear the importance of the bal-
ance between the need for diagnostically useful images and the need
to keep radiation exposures as low as reasonably achievable. This
clarification would not have been feasible without the earlier
preparatory work and it meant that the research programme was
facing up to the basic trade-off central to the acceptance and full

7
Medecine.EN (Flash) 23/10/02 16:34 Page 8

implementation of radiation protection principles in diagnostic
radiology. This basic objective was supported by more technical
objectives on development of methods to parameterise the radio-
logical information content and to adapt strategies from adult to
paediatric patients.
This general thrust was maintained into FP5. The overall objective
for this period under a general programme on medical and indus-
trial uses of radiation was further development of methods for optimising
the use of radiation in medicine. This was supplemented by investi-
gation of new diagnostic techniques that offer advantages in terms
of cost, diagnostic information and lower exposures.
In view of the relatively small number of projects and the conti-
nuity between FP4 and FP5 in this area the research projects and
results achieved are considered together for the two programmes.
Summaries of all the projects are provided in Part 2 of this brochure.
Quality assurance and quality criteria
Projects concerned with specific aspects of quality assurance and
quality criteria formed part of FP4. These were aimed at the com-
pletion and enhancement of the collected results of the previous
programmes and resulted in publication of the four volumes on
Quality Criteria [3-6] (see Box 3).

Box 3 : Major publications from the research programmes

European Guidelines on Quality Criteria for Diagnostic Radiographic Images: EUR 16260 EN June 1996

European Guidelines on Quality Criteria for Diagnostic Radiographic Images in Paediatrics: EUR 16261 EN June 1996

European Guidelines on Dosimetry in Mammography: EUR 16263 June 1996

European Guidelines on Quality Criteria for Computed Tomography: EUR 16262 EN May 1999

The optimum use of diagnostic radiology involves the interplay of Visualization: Characteristic features are detectable but details
three important aspects of the imaging process: are not fully reproduced; features just visible.

The diagnostic quality of the radiographic image
The radiation dose to the patient, and
The choice of radiographic technique.
The four documents on Quality Criteria and Dosimetry shown in
Box 3 provide guidance on all three aspects. The diagnostic qual-
ity is represented by image criteria for different types of radi-
ographs that were found by surveys and trials to be those deemed
necessary to produce an image of standard quality.
Typically these are mainly qualitative descriptors using phrases such as:
Reproduction: Details of anatomical structures are visible but not
necessarily clearly defined; details emerging.
Visually sharp reproduction: Anatomical details are clearly defined;
details clear.
These are accompanied where appropriate by a quantitative indi-
cation of the resolution of important image details.
For each projection there are criteria as to which elements have to
satisfy the various descriptors. See for example Box 4.

Box 4 : Quality criteria for Computed Tomography

Lumbar Spine Lateral Projection [6].

1. DIAGNOSTIC REQUIREMENTS
1.1. Image criteria
1.1.1. Visually sharp reproduction, as a single line, of the upper and lower-plate surfaces with the resultant visualization of
the intervertebral spaces
1.1.2. Full superimposition of the posterior vertebral edges
1.1.3. Reproduction of the pedicles and the intervertebral foramina
1.1.4. Visualization of the spinous processes
1.1.5. Visually sharp reproduction of the cortex and trabecular structures
1.2. Important image details: 0.5 mm.
2. CRITERIA FOR RADIATION DOSE TO THE PATIENT
Entrance surface dose for a standard-sized patient: 30 mGy

8
Medecine.EN (Flash) 23/10/02 16:34 Page 9
With respect to criteria for radiation dose to the patient, these are
given as reference dose values. These values have been derived
from large-scale European surveys carried out in earlier research pro-
grammes and at national level, and are set at the 3rd quartile (75th
percentile). They are intended to initiate immediate investigation
if exceeded. Otherwise they should be seen as the ceiling from
which progress should be pursued to lower dose levels in line with
the optimisation principle. The reference dose values are given in
the appropriate units for the relevant measurement quantity for the
type of examination.
Finally the examples of good radiographic practice have resulted
from evaluation of the results of European trials of earlier versions
of the quality criteria.
Although it was decided to complete the set of quality criteria
guideline documents issued in 1996 with one on CT and this was
one of the main tasks of a project under FP4 (page 18), it was clear
that in such a complex and fast developing field unanswered ques-
tions remained. In particular in comparison with conventional radi-
ology, the complexity, range and flexibility of scanner settings in
CT may adversely affect the levels of image quality and patient dose
achieved in practice. The initial sets of criteria that were published
in 1999 [6] need to be thoroughly “road-tested”. Emerging tech-
niques such as multislice CT and fluoro-CT will lead to a need for
review and updating of the guidelines.
In developing and building on these guidelines two projects were
concerned with the particular problems of paediatric examinations
(page 26, 28). In particular, given the high-dose examinations car-
ried out on children using fluoroscopy and CT, there was a need to
extend the guidelines for adults to cover children. As a result of the
research programme sets of quality criteria were derived specific
to the anatomy and pathology of paediatric patients of 0, 1, 5, 10
and 15 years old. Factors were derived for normalising both entrance
surface dose and dose-area product measurements. The optimum
choice of technique was also investigated taking into account the
particular requirements of image detail for children and examining
the causes of rejection of images as unacceptable. All of this work
is now available to contribute to further elaboration of the set of
quality criteria guidelines.
The radiation dose to the patient has to be measured in a repro-
ducible and consistent manner. To assist with this, one project
(page 30) dealt with the criteria for laboratories providing dosimetric
services in the context of diagnostic radiology. In particular the pro-
ject was aimed at developing recommendations for patient dosime-
try using thermoluminescent dosimeters (TLDs). The final document
published by the Commission in 2000 [7] covered both the con-
ceptual and practical aspects of patient dosimetry in general diag-
nostic radiology, mammography and CT. The quantities to be
measured included entrance surface dose, dose area product and
computed tomography dose index.
Dental radiography is an extremely widespread application of
diagnostic imaging. Although the estimated risk from exposure in
dental radiography is relatively small in terms of other risks, its fre-
quent use and the rapid technological development warranted
some research into optimisation and quality assurance. Through this
project (page 24) it was possible to develop quality criteria and pre-
PART 1
sent information on the technical and clinical performance of state
of the art digital imaging techniques. These were used to recom-
mend optimisation strategies that have been promulgated in a
number of specialist publications.
Development of new technologies
Screen film radiography is still the primary means of x-radiography.
However the benefits of a transition to digital radiography include
the higher radiation collection efficiency and the larger dynamic
range of new detectors. These benefits could be significant in terms
of departmental productivity as well as for diagnostic quality and
dose reduction. For this reason a project (page 22) to investigate
the potential of these new detector designs was included in FP4.
The primary research emphasis was on active matrix flat panel
imagers using amorphous silicon arrays and on other designs of
hybrid solid state semiconductors. All of these products have poten-
tial but there is a long transition between improved detector per-
formance in the laboratory and improvements in clinical practice.
Similarly the potential for dose reduction may not be realised unless
this is kept in mind during the whole process of prototype design,
testing and production engineering. From this point of view the pro-
ject has hopefully helped to alert developers and manufacturers to
the multiple objectives they must keep in view.
Optimisation of radiological information and patient protection
This most substantial objective of the entire medical radiation pro-
tection research programme has been expanded from one project
in FP4 into three projects in FP5. In broad terms work in FP4 con-
centrated on screen-film as the imaging medium; work in FP5 is con-
centrating on applying the experience gained to the assessment of
image quality produced by digital imaging systems.
A major part of the work in this area is represented by the DIMOND
Concerted Action (Digital Imaging: Measures for Optimising
Nationally radiation Dose), which started in FP4 (page 12) and con-
tinued with a different co-ordinator in FP5 (page 14). The research
was directed at selected procedures involving computed radiogra-
phy with storage phosphor systems, image intensifier radiography
and interventional radiography. The problems studied were those
presented by changing population exposures, by the more detailed
analysis of radiological information content, and by the joint opti-
misation of dose and image quality. A number of subjective and
objective indices of image quality were investigated. Subjective
indices, which were those used almost exclusively in the three pub-
lications on Quality Criteria, have the advantage of mimicking the
actual detection tasks and are reasonably simple to use but are nec-
essarily imprecise and difficult to maintain consistently between cen-
tres. Objective indices such as signal to noise ratio are more complex
to measure but provide more consistent results. The question of how
closely these indicators correlate with subjective quality assess-
ments is still under investigation in FP5 concentrating on digital imag-
ing for interventional studies and mammography.
As mentioned earlier, the Quality Criteria for CT established during
FP4 were necessarily preliminary. To refine them further research
under FP5 is being carried out by a consortium known as the
European Working Group on Computed Tomography (page 16). This
work is also needed to respond to the new Council Directive [8], which
9
Medecine.EN (Flash) 23/10/02 16:34 Page 10
puts specific requirements on quality assurance and patient dose
assessment in CT. To provide up to date data from which to derive
image quality and radiation protection indicators a survey is being
carried out for 10 relevant clinical indications for CT examination for
skull, chest and abdomen that involves hospitals in 5 countries or
more. The outcome will be a revision and extension of the document
EUR 16262 [6] to be prepared at the end of the contract period.
The most challenging project under FP5 is concerned with the unifi-
cation of physical and clinical requirements for x-ray imaging using
digital techniques (page 20). This is also attempting to develop more
objective assessment methods employing visual detection techniques.
Those being employed are visual grading and receiver operating char-
acteristic. This approach has highlighted the role of the viewer under-
taking the visual assessment in the overall scheme and the development
of relationships is at an advanced stage. A set of physical descriptors
for defining imaging performance is also being developed.
In combination, all these projects involve a very large group of
contractors, including many of Europe’s main medical research
organisations, with a good balance between clinical, physical and
technological skills and experience. Significant cost savings have been
achieved through the sharing of understanding and the spreading
of measurements throughout representative facilities in many coun-
tries. This, and the knowledge and experience built up during pre-
vious research collaborations, has enabled such a wide ranging
problem to be tackled fully for the first time. Participation has
served to consolidate and improve the focus of national research
programmes and the multinational interactions have contributed
to the widespread adoption of results. It is expected that the results,
reflecting best practice in a European context, will become avail-
able as updates of the Quality Criteria Guidelines appropriate for
the new digital imaging techniques and incorporating objective cri-
teria to supplement the subjective criteria used until now.
5. Priorities for the 6th Framework Programme
Radiation protection is one of the thematic priorities in the next
Euratom Research Programme and as medical uses of radiation are
such a dominant contributor to the radiation exposure of the
European population this area will continue to be of importance.
The exact nature of the research priorities and the resources to be
allocated has yet to be decided and these decisions will involve input
from many “stakeholders”. Nonetheless, a number of issues have
already emerged on which further Community research in the 6th
Framework Programme (FP6) would appear to be warranted.
In general, based on the most recent survey by UNSCEAR [9], fur-
ther increases in the uses of medical radiations and resultant doses
can be expected facilitated by advances in health care technology
and by economic developments. The latter will be particularly
important in a European context as Eastern European countries join
the Community and endeavour to reach common European stan-
10
dards of both health care and associated patient protection. Increases
are to be expected in the utilisation of x-rays, with a continuing
growth in importance for CT partly fuelled by the introduction of
new techniques such as multislice and fluoroscopic CT. Digital imag-
ing will also increasingly replace film and enable an expansion of
direct interventional procedures with their potential for high doses
to both patients and staff. Practice in nuclear medicine will be dri-
ven by the use of more specific radiopharmaceuticals for both diag-
nosis and therapy, and there will be an increasing demand for
radiotherapy owing to population ageing.
As exemplified by the changes in emphasis between FP4 and FP5
from film to digital imaging techniques, there will be a continuing
need to carry out research to ensure that application of radiation
protection principles in medical diagnostic radiology remains appro-
priate and that Quality Criteria for both imaging and patient pro-
tection are developed and applied together. The growth of CT as
an imaging modality from nothing to one of the more common tech-
niques in less than 3 decades shows how rapidly the medical pro-
fession can exploit technical advances.
There has been relatively little research at Community level into
patient protection in nuclear medicine over the past decade. During
this period many new radiopharmaceuticals and procedures have
come into use and new imaging techniques are being developed.
It would appear timely during FP6 to look again at developments
in this area to ensure that protection measures are keeping pace
with technical developments.
Radiation protection in therapy has not previously been a signif-
icant component of the Commission’s research programmes. This
probably stems from the fact that the primary objective is to
deliver the correct dose to the tumour rather than try to minimise
it. However as this primary objective is increasingly achieved and
survival rates increase, more attention should be paid to minimi-
sation of doses to surrounding tissues so that the occurrence of
secondary radiation-induced tumours can also be kept to a mini-
mum. It is also being recognised that radiosensitivity differs
between individuals. This means that using generalised treatment
patterns may result in too much or too little exposure of both the
tumour and surrounding tissue with implications for treatment suc-
cess and for radiation protection. To take this into account in rou-
tine operations may require the application of underlying
radiobiological research in a clinical context. A different problem
is the prevention of “accidents” which in the radiotherapy con-
text means the delivery of significantly more or less dose to the
tumour than specified. Either of these is potentially fatal and
there have been an increasing number of such accidents reported
by the IAEA recently [10]. The means of prevention is principally
through excellence in quality assurance coupled with a good
safety culture among the staff. Research into the best ways to
achieve such excellence could form the basis for stronger
Commission actions in this area.
Medecine.EN (Flash) 23/10/02 16:34 Page 11

PART1

4th Programme (1994-1998)

(1) Radiation protection of the patient in paediatric
radiology

(2) Digital imaging: Measures for optimising radiological

information content and dose (DIMOND II)

(3) Quality criteria for Computed Tomography (QCCT)

(4) Optimisation of radiation protection of the patient

in paediatric radiology (ORPED)

(5) Criteria and protocols for central laboratories

providing dosimetric services for quality control
in diagnostic radiology (DIAGNOSDOS)

(6) New detector techniques in x-ray diagnosis and

impact on radiation protection strategies

(7) Maxillary imaging network on optimisation strategies

and quality assurance of radiographs in Europe
(MINOSQUARE)

5th Programme (1998-2002)

(8) Unification of physical and clinical requirements for

medical x-ray imaging and its relevance to European
industrial and socio-economic development
(UCP X-IM)

(9) Measures for optimising radiological information

and dose in digital imaging and interventional
radiology (DIMOND III)

References (10) Computed tomography – techniques, image quality

and patient dose (CT-TIP)
[1] Council Directive 84/466/EURATOM
[2] 1990 Recommendations of the International Commission on
Radiological Protection. ICRP Publication 60. Annals of the ICRP Vol. 21
No. 103, 1991
[3] European Guidelines on Quality Criteria for Diagnostic Radiographic
Images: EUR 16260 EN June 1996
[4] European Guidelines on Quality Criteria for Diagnostic Radiographic
Images in Paediatrics: EUR 16261 EN June 1996
[5] European Guidelines on Dosimetry in Mammography: EUR 16263
June 1996
[6] European Guidelines on Quality Criteria for Computed Tomography:
EUR 16262 EN May 1999
[7] Recommendations for Patient Dosimetry in Diagnostic Radiology using
TLD. Nuclear Science and Technology Topical Report. EUR 19604 EN,
2000
[8] Council Directive 97/43/EURATOM
[9] Sources and Effects of Ionizing Radiation. UNSCEAR 2000 Report to
the General Assembly. United Nations Scientific Committee on the
Effects of Atomic Radiation. New York, 2000
[10] Lessons Learned from Accidental Exposures in Radiotherapy.
International Atomic Energy Agency. Safety Reports Series No. 17
STI/PUB/1084, 2000

11
Medecine.EN (Flash) 23/10/02 16:34 Page 12

bjectives

The primary objectives of the

DIMOND Concerted Action con- Digital imaging: measures for optimising
cern the problems and challenges
presented by changing population
radiological information content and dose
exposures, parameterisation of (DIMOND II)
radiological information content,
and dose and image quality optimi-
sation for selected procedures in
Computed Tomography (CT).
These procedures involve storage
phosphor systems, image intensi-
fier radiography and interventional
radiography. To address these
problems the project includes sub-
stantial reviews to ascertain the
broad current scientific consensus
on the measurement, interpreta-
tion and practical implementation
of the relevant aspects of patient
and staff dosimetry, image quality
and radiological information con-
tent. The development of measure-
ment protocols and methodologies
is intended to be a significant out-
come of the project.
The more specific objectives are
related to the review of methods Achievements Modular Transfer Function (MTF) and
and protocols for the assessment of Wiener Spectra (WS) as important and
radiological image content / image The selected procedures for the quality cri- widely used objective indices of image qual-
quality for particular applications, teria study were chest and skeletal imaging ity, even though their interpretation is also
and for patient dose measurement. for storage phosphor radiography, small not well understood.
bowel examinations and ascending venog-
raphy of the legs for image intensifier radi- An interim review has been carried out
ography, and angiography of the pelvis and specifically for DSA. A number of currently
lower extremities and the dilation of the iliac used measures of image quality for DSA
arteries for Digital Subtraction Angiography have been cited for both subjective and
(DSA). An interim review has been conducted objective approaches. Here again MTFs and
of the clinical visualisation criteria for these WSs are useful indices. There appears how-
procedures to categorise the image criteria, ever to be no current consensus on optimum
important image details and critical struc- dose levels for DSA, furthermore DSA opti-
tures. These criteria have been developed misation would appear to be driven by clin-
through informed consensus opinion, and ical criteria rather than the technical
formed an input to the Guidelines on CT limitations of present technology.
[European Guidelines on Quality Criteria for
Computed Tomography EUR 16262]. Constancy check methodologies have been
reviewed for storage phosphor systems,
Methods and protocols for assessment of coronangiography and DSA systems, and
radiological information content and image also dosimetry equipment. These reviews
quality have been reviewed for particular indicated the advantages to performing
digital applications. It is apparent from this constancy checks, the features of different
work that a broad consensus indicated that equipment that must be checked, and the
the objective index of Signal-to-Noise Ratio components of constancy check methodol-
(SNR) is related to radiological information ogy that are practices by all participants in
content, but details of this relationship the concerted action. These reviews will be
remain vague among the scientific commu- used to develop protocols for the different
nity at large. This project also identified equipment classes.

12
Medecine.EN (Flash) 23/10/02 16:34 Page 13

PART 2
DIMOND II Information Column

Various approaches to dosimetry in digital,
fluoroscopic and interventional systems have
been reviewed, and three dosimetric
approaches identified. These are 1) use of
an ionisation chamber to determine patient
entrance skin dose; 2) monitoring of dose
area product; and 3) use of TLDs placed on
patients’ skin at the centre of the field. Other
progress involved the completion of a TLD
intercomparison among the participants.
Finally there has been some progress
regarding the definition of equipment
requirements in the form of a draft docu-
ment entitled “Requirements for Dose
Display Systems Integrated in Radiological
Equipment”. This document states that dis-
play systems should ideally provide Effective
Dose as well as Entrance Surface Dose.
Title:
An international Workshop was organised in Digital imaging: measures for
Dublin in June 1999 entitled “Dose and Image optimizing radiological information
Quality in Digital Imaging and Interventional content and dose (DIMOND II)
Radiology” in order to present the results of
Co-ordinator:
this Concerted Action. The proceedings are
James F. Malone
in the process of being published. Federated Dublin Voluntary Hospital
Department of Medical Physics
and Bioengineering
Garden Hill House, St. James’s Hospital
Outlook IRL-DUBLIN 8,
Ireland
While this Concerted Action group has Tel: +353.1.453.7941
made significant progress during the con- Fax: +353.1.473.2712
tract period, there is still a substantial e-mail:
body of research to be completed arising jfmalone@haughton-institute.ie
from the progress, achievements, and
Partners:
deliverables in terms of clinical, techni- - Keith Faulkner
cal, industrial, standards and regulatory (Quality Assurance Reference Centre,
objectives. Newcastle-upon-Tyne, UK)
- Hans-Peter Busch
(Brüderkrankenhaus, Trier, Germany)
- S. Rannikko
(STUK, Helsinki, Finland)
- G.J.F. Marchal
(KU Leuven, Belgium)
- Werner Jaschke
(University Hospital Innsbruck,
Austria)
- Carlo Back
(Ministry of Health, Luxembourg)
- Eliseo Vañó
(Universidad Complutense
de Madrid, Spain)
- V. Neofotistou
(Athens General Hospital, Greece)
- Johannes Zoetelief
(TNO, Delft, The Netherlands)
- Renato Padovani
(Ospedale Santa Maria della
Misericordia, Udine, Italy)

EC Scientific Officer:
Diederik Teunen
(Contact: Ernst-H.Schulte)
Tel: +32.2.295.71.55
Fax:+32.2.295.49.91
e-mail:
ernst-hermann.schulte@cec.eu.int

Period:
Nuclear Energy 1994-1998

Status:
Completed

13
Medecine.EN (Flash) 23/10/02 16:34
bjectives
The medical use of radiation in
interventional radiology, digital
imaging and mammography offers
enormous potential benefits to
patients but involves some risks.
Interventional radiology is
increasingly popular because some
patients can be treated as
outpatients where the surgical
alternative involves long
hospitalisation and is more
traumatic for the patient.
The objective of the project is to
achieve safer, more cost-effective
health care using these procedures.
To do so it is necessary to assess the
radiation doses and risks to
patients, so as to improve the
techniques used in examinations.
This will be achieved by investigating
methods for the optimisation of
image quality and dose for specific
clinical examinations.
14
Page 14
Measures for optimising radiological
information and dose in digital imaging
and interventional radiology (DIMOND III)
Challenges to be met
An interventional procedure is one in which
radiological imaging equipment is used to
guide a doctor in the placement of a med-
ical device, either for diagnostic or thera-
peutic purposes. Interventional procedures
have become common in cardiology and
many other major areas of medicine.
A digital mammogram taken on a full field
digital system.
The procedures, which can involve long flu-
oroscopy times, often require the placement
of a wire, for example via the femoral artery,
into the heart, or brain, by gradual move-
ments. The wire is difficult to see on the
live television image and to manipulate it to
a specific position requires great dexterity.
These procedures present a number of dif-
ficult radiation protection problems.
The geometry of the x-ray tube and image
intensifier lead to high dose rates from scat-
tered radiation in the vicinity of the patient
couch where the doctor and assistants nor-
mally stand. If there is a lack of awareness
of radiation protection, dose levels to both
patients and staff can be higher than nec-
essary. Indeed deterministic injuries, such as
epilation, radio-dermatitis and skin necro-
sis have been observed in patients who
have had interventional procedures.
The female breast is particularly radiosen-
sitive. It is therefore important to evaluate
the risk/benefit ratio for digital mammog-
raphy if it is to be used for screening pur-
poses. If a cancer is suspected the patient
will be referred for further assessment,
which may involve the accurate placing of
a needle within the breast to take a sample
of cells or a small piece of tissue.
Digital imaging will improve the accuracy
of needle placement and speed up the pro-
cedure. However it is necessary to ensure
that the procedure and dose are optimised
so that the woman does not receive an
excessive radiation dose.
The specific investigations to be carried out
under this project are:
◗ to find methods of improving image qual-
ity without increasing doses;
◗ to optimise the image quality/dose rela-
tionship;
◗ to develop radiological image content and
quality criteria;
◗ to produce equipment requirements and
specifications;
◗ to investigate dosimetry approaches for
both patients and staff; and
◗ to find methods of presenting the doctor
carrying out the procedure with under-
standable dosimetric information.
Achievements
The evaluation of medical imaging systems
may be performed using both subjective and
objective indices. Subjective indices, such as
the detection of objects mimic the actual
detection tasks and are simple to use but are
an imprecise approach. Objective indices are
harder to measure but provide more detailed
information on image quality. Measurements
using both types of indices have been per-
formed on various digital systems.
Medecine.EN (Flash) 23/10/02 16:34 Page 15
A photograph of a cardiologist’s eye showing:
1) posterior sub-capsular opacity;
2) paranuclear dot opacities.
Requirements and specifications for the pur-
chase of digital interventional radiology equip-
ment have been developed and published.
Constancy check protocols for the image qual-
ity and dosimetric assessment of fluoroscopy
equipment have also been developed.
A case of radio-dermatitis in a 17 year old
patient, two years after two cardiac ablation
treatments (Vano et al. 1998).
Initial studies into the optimisation of dig-
ital imaging equipment have commenced.
As the dose is a user-selectable variable, it
is important to select the dose level that
yields the image quality necessary for the
examination. Quality criteria studies have
begun on digital imaging equipment for
interventional studies and mammography.
PART 2
DIMOND III Information Column
Title:
Measures for optimising radiological
information and dose in digital
imaging and interventional radiology
(DIMOND III)
Co-ordinator:
Keith Faulkner
Quality Assurance Reference Centre
Newcastle General Hospital
Westgate Road
UK-NE4 6BE Newcastle upon Tyne,
United Kingdom
Tel: +44.191.256.3882
An ongoing study is underway to develop a Fax: +44.191.219.5034
complexity index for cardiac interventional e-mail:
procedures. This has involved a study of the keith.faulkner@ncht.northy.nhs.uk
relationships between clinical and technical
Partners:
factors. Clinical factors investigated include - James F. Malone (Haughton Institute
age, sex, ejection traction and a previous for Graduate Education and Training
coronary artery bypass graft. Technical fac- in the Health Sciences Ltd., Dublin,
tors include multi-vessel angio-plastics, use of Ireland)
double-wire or balloon type of stenting and - Hans-Peter Busch
intravascular ultrasonography. (Brüderkrankenhaus, Trier,
Germany)
In mammography, studies into the risk/ben- - Eliseo Vañó
efit ratio and measures for quality assur- (Hospital Clinico San Carlos and
ance have begun. Preliminary results indicate Complutense University, Madrid,
that breast screening is justified in radia- Spain)
tion protection terms. - Hilde Bosmans (Katholieke
Universiteit Leuven, Belgium)
- Werner Jaschke (University Hospital
Partnership Innsbruck, Austria)
- Renato Padovani
The DIMOND III group comprises of 13 part- (Azienda Ospedaliera Santa Maria
ners from 11 member states. It brings together della Misericordia, Udine, Italy)
a group of medical specialists from a wide - Alexandra Schreiner
range of backgrounds. Most partners are (Centre de Recherche Public-Santé,
based in hospitals but the group includes Luxembourg)
partners with academic and legislative back- - Antti Servomaa
grounds. This variety of backgrounds pro- (STUK, Helsinki, Finland)
vides a unique breadth and depth of expertise. - Johannes Zoetelief (Delft University
of Technology, The Netherlands)
- Michael Molfetas (Evagelismos
General Hospital, Athens, Greece)
- John Kotre (Newcastle General
Hospital, UK)
- Sylvia Vetter
(Evangelische Diakonissenanstalt
Karlsruhe, Germany)
EC Scientific Officer:
Ernst-H. Schulte
Tel: +32.2.295.71.55
Fax:+32.2.295.49.91
Development of a complexity index for cardiac
e-mail:
interventional procedures.
ernst-hermann.schulte@cec.eu.int
Period:
Nuclear Energy 1998-2002
Status: Ongoing
15
Medecine.EN (Flash) 23/10/02 16:34 Page 16

bjectives

Computed tomography (CT) is a

procedure that involves relative
high radiation exposure of the
Computed tomography – techniques,
patient and for which dosimetric image quality and patient dose (CT-TIP)
field surveys have revealed large
variations in exposures between
hospitals. The objective of the
European Working Group on
Challenges to be met become more involved in radiation protec-
Computed Tomography (EWGCT) tion, in the optimisation of CT protocols
is to stimulate optimisation of In Europe, CT is estimated to be responsi- and in clinical patient dosimetry. This is also
applications of CT. To achieve this, ble for up to 40% of the collective dose due partly due to recent European legislation,
insight in the clinical application to diagnostic radiology. In recognition of which puts specific requirements on quality
of CT in European countries will be this, professional groups, governments and assurance programmes in CT and on the
manufacturers as well as individual medical assessment of patient dose in CT (Council
obtained and further guidance on
specialists, radiographers and physicists have Directive 97/43/EURATOM).
imaging technique, patient dose
and diagnostic accuracy will be
provided. Information will be col-
lected with regard to CT technique
and to patient dose in relation to a
large variety of clinical problems.
A European evaluation of image
quality will be performed for a
limited number of CT procedures.
New data for calculating effective
doses for new types of CT scanners
will be provided. Finally, the new
insights will be incorporated in a
revision of the European Quality
Criteria for CT.

Figure 2 – Flowchart for computed tomography.

A radiologist might decide to use CT as the
most appropriate diagnostic imaging
modality. In this case it is expected that the
diagnostic accuracy of the CT scan will
contribute to the health benefit of the
patient, e.g. through treatment of the
detected disease. The radiation exposure
associated with the CT scan will however
give rise to a health detriment, i.e. an
increased probability of tumour induction
or hereditary effects. After clinical justifi-
Figure 1 – A computed tomography (CT)
cation a potential for optimisation of CT
scanner.
can often be found in the actual imaging
technique, especially in the choice of tube
current, pitch, number of sequences and
volume of the scan. The imaging technique
determines to a large extent the absorbed
dose and the diagnostic accuracy. It is a
challenge for the EWGCT to achieve opti-
misation of imaging techniques in CT by
stimulating structured evaluations of
absorbed dose and diagnostic accuracy.
Recent developments in CT have been
fuelled by new technologies such as sub-
second multislice CT and fluoroscopic CT
and by novel clinical insights. It is another
challenge for the EWGCT to incorporate
these developments in their new recom-
mendations.

16
Medecine.EN (Flash) 23/10/02 16:34
Figure 3 – The European guidelines on
quality criteria for computed tomography
(EUR 16262 EN).
Achievements in the
4th Framework Programme
The point of departure for the concerted
action is the European Guidelines on
Quality Criteria for CT examinations that
have been established within the 4th
Framework Programme. The objectives of
the European Guidelines are to achieve
adequate image quality and reasonably
low radiation dose per examination. The
Quality Criteria define a level of perfor-
mance considered necessary to produce
images of standard quality for a specific
clinical problem. The criteria involve exam-
ples of good imaging technique, criteria for
radiation dose to the patient and diag-
nostic requirements. Although they were
a source of much interest in Europe, they
are not yet widely implemented.
Expected results of the
5th Framework Programme
This project aims at stimulating the optimi-
sation of CT practice in Europe with regard
to image quality and patient dose and in
relation to the specific clinical problem of the
patient. Several results that are expected
will help to achieve these objectives.
Page 17
PART 2
CT-TIP Information Column
Title:
A European field survey of the clinical appli- Computed tomography – techniques,
cation of CT with a focus on the evaluation image quality and patient dose
of CT protocols and assessment of patient (CT-TIP)
dose will be organised. This survey is based
Co-ordinator:
on more than 10 relevant clinical indica-
Jacob Geleijns
tions for CT examinations and comprises Leiden University Medical Center
scans of the skull, chest and abdomen. The P.O. Box 9600
survey involves four hospitals in at least 5 Albinusdreef 2
countries. The results of the survey will be NL-2300 RC LEIDEN,
evaluated with regard to optimisation of CT The Netherlands
technique and patient dose. Results will also Tel: +31.71.526.20.49
be relevant for the further development of Fax: +31.71.524.82.56
the diagnostic reference dose levels in e-mail: j.geleijns@lumc.nl
Europe and for the revision and extension
of the European Guidelines. Partners:
- Georg Bongartz
(University Hospital-Kantonspital
Subsequently a field survey focused on Basel, Switzerland)
image quality in CT for a selected group of - Stephen J. Golding
clinical problems and hospitals will be (Oxford MRI Centre, John Radcliffe
organised. From the results of the broad Hospital, UK)
survey described some indications will be - Karl Arne Jessen
selected for further evaluation. This selec- (Aarhus Kommunehospital,
Denmark)
tion will be based on the variations in the
- Marco Leonardi
CT protocols or in patient dose that are
found during the broad survey of the stan- (Ospedale Bellaria, Bologna, Italy)
dard protocols or on the deviations from the - Ricardo Rodriguez Gonzalez
(Hospital Clinico Universitario
‘Examples of Good Imaging Technique’ in
San Carlos, Madrid, Spain)
the European Guidelines. The evaluation - Karl Schneider
involves patient dose but a special effort will (Dr. von Haunersches Kinderspital,
also be made in evaluating clinical image Munich, Germany)
quality. To achieve this, the set of collected - Werner Panzer
clinical CT images are to be evaluated dur- (GSF Neuherberg, Germany)
ing a joint session of the radiologists. - Paul Shrimpton
(NRPB, Didcot, UK)
A revision and extension of the CT Quality - Alfonso Calzado Cantera
Criteria aimed at the preparation of a gen- (Universidad Complutense
eral update of the Guidelines will be pre- de Madrid, Spain)
- Giampiero Tosi
pared at the end of the contract period. Rapid
(Istituto Europeo di Oncologia,
development of clinical applications of CT Milan, Italy)
will necessitate a revision of this document.
EC Scientific Officer:
A method for characterising the dosimet- Ernst-H.Schulte
ric performance of the latest types of CT Tel: +32.2.295.71.55
scanner will be developed so that appro- Fax:+32.2.295.49.91
priate conversion factors can be derived e-mail:
and effective doses calculated for current ernst-hermann.schulte@cec.eu.int
clinical CT practice. Special attention will
be paid to the evaluation of the relation Period:
between dose length product (DLP) and Nuclear Energy 1998-2002
effective dose. The results of these studies
Status:
will be applied for dose assessment during Ongoing
the two field surveys.
17
Medecine.EN (Flash) 23/10/02 16:34 Page 18

bjectives

Computed Tomography (CT) is

recognised as a high dose imaging
modality, contributing up to 40%
of the collective effective dose from
diagnostic radiology in some EU
countries. The quality criteria
concept has proved to be an
efficient means for optimising the
use of ionising radiation in medical
imaging procedures involving
conventional and paediatric
radiology. The objective was to
establish quality criteria for CT
examinations that are relevant both
to the provision of clinical
information in an optimum form
and also to the control and potential
reduction of dose.
Picture 1 – Document EUR
16262 with references.
Quality criteria
for computed tomography (QCCT)
Challenges to be met
In 1994 a Study Group was established and a Working
Document on Quality Criteria for CT was produced.
This was sent for consultation to outside experts,
competent authorities and professional bodies in the
Member States. It has been one of the main tasks of
the project to analyse, unify and incorporate over 50
responses to produce a final document. In parallel
with this consultation, trials of the procedures in the
working document have had to be carried out in
clinical departments. One of the proposals in the
Working Document was to use Dose-Length- Product
(DLP) as a method for specifying reference doses for
CT. This proposal had to be evaluated. A means of
allowing ready access to technical information had
to be established and finally the results of the pro-
ject had to be disseminated.

Achievements

The usability in clinical practice of the image
quality criteria proposed in the working doc-
ument was evaluated in a pilot multicentre
study for five types of examinations 1) face
and sinuses, 2) vertebral trauma, 3) HRCT of
the lung, 4) liver and spleen, and 5)
osseous pelvis. CT examinations from four
countries were evaluated by two groups of
radiologists. For each examination the radi-
ation dose was evaluated based on
Computed Tomography Dose Index (CTDI)
measurement in the centre and periphery of
a CT dosimetry phantom and recorded as
DLP. It was concluded that the highest dose
does not always imply the best diagnostic
image and that dose reduction could be
achieved without loss of diagnostic image
quality. These results as discussed at a work-
shop in Malmo in June 1999 have been pub-
lished (Rad. Prot. Dos. 90, 47-52, 2000).
The CTDI is a well-defined and useful dose
descriptor of the dose to the patient from
CT. It is easy to measure and thus practica-
ble for defining reference doses. It has been
recommended by the International Electro-
technical Commission that values of CTDI
should be displayed on the operator’s con-
sole of the CT scanner, reflecting the con-
ditions of operation selected. Monitoring of
CTDI for the head or body CT dosimetry
phantom provides control on the selection
of exposure settings, such as mAs.
Monitoring of the DLP where the number
of slices and sequences are taken into
account provides control on the volume
irradiated and the exposure from the irra-
diation as a whole. During the project a
pilot study provided some preliminary data
on appropriate values for DLP. These data
together with some from a separate UK
survey enabled initial reference dose values
to be proposed; however these will need
further checking by means of a wider trial.
Effective dose to the patient can be calcu-
lated from the operational dose quantities
CTDI or DLP thus enabling the different
examinations to be compared meaningfully
taking into account the non-uniform expo-
sure of the body.
A reference database for CT dosimetry has
been established at ImPACT in the UK. This
can be accessed at www.impactscan.org/.

18
Medecine.EN (Flash) 23/10/02 16:34 Page 19

PART 2
QCCT Information Column

Title:
Quality Criteria for Computed
Tomography (QCCT)

Co-ordinator:
Karl Arne Jessen
Aarhus University Hospital
Noerrebrogade 44
DK-8000 AARHUS C,
Denmark
Tel: +45.894.92.591
Fax: +45.894.92.590
e-mail: kaj@medfysic.aau.dk

Partners:
- Anne Grethe Jurik
(Aarhus University Hospital, Denmark)
- Georg Bongartz
(University Hospital-Kantonspital
Basel, Switzerland)
- Jacob Geleijns
(Leiden University Medical Center,
The Netherlands)
- E. v. P. van Meerten
(Leiden University Medical Center,
Picture 2 – A CT-scanner installation. The Netherlands)
- Stephen J. Golding
(Oxford MRI Centre, John Radcliffe
Hospital, UK)
- Marco Leonardi
Implications (Ospedale Bellaria, Bologna, Italy)
- Werner Panzer
A workshop was organised in November This initiative in CT should continue with (GSF Neuherberg, Germany)
1998 in Århus, Denmark to discuss all of the a view to ensuring compliance with the - Paul C. Shrimpton
responses to the Working Document. There Council Directive on health protection of (NRPB, Didcot, UK)
were no fundamental critical comments but individuals against dangers of ionising - Giampiero Tosi
some discussion was needed to arrive at a radiation in relation to medical exposure (Istituto Europeo di Oncologia,
general agreement to and support for the (97/43/EURATOM). The quality assurance Milan, Italy)
application of the dosimetric quantities CTDI criteria required by the Directive and
and DLP, which were introduced into the established in preliminary form during EC Scientific Officer:
document. The document was then finalised this project should be tested by regular Diederik Teunen
(Contact: Christian Desaintes)
as an official publication of the Commission repeated surveys which include the appro-
Tel: +32.2.295.82.73
[European Guidelines on Quality Criteria for priateness of the initially selected refer-
Fax:+32.2.295.49.91
Computed Tomography, EUR 16262 EN, May ence dose values. Emerging techniques e-mail: christian.desaintes@cec.eu.int
1999]. This document can also be found at such as multislice and fluoro-CT have not
www.drs.dk/guidelines/ct/quality. been addressed in this project and with Period:
the continuing evolution of CT technology Nuclear Energy 1994-1998
there will be a need for regular updating
of the guidelines. Status:
Completed

Picture 3 – Dose measurements in a phantom.

19
Medecine.EN (Flash) 23/10/02 16:34
bjectives
The ability to quantify the image
quality outcome of medical x-ray
examinations is fundamental to our
understanding of the role and
relevance of diagnostic radiology in
modern health care. Indeed the
discovery of x-rays heralded the
beginning of today’s
industrial/technological framework
for modern health-care systems.
The primary objective of this project
is to develop methods for assessing
the quality of radiographic images
from the clinical anatomical
structures that are presented,
whether normal or abnormal.
This work builds upon previous
European led research in the
application of visual detection
techniques which not only study
what contributes to acceptable
image quality but also the role
played by the person who clinically
assesses medical images.
With the advent of new digitally
based x-ray imaging technology
the time is ripe to tackle this
problem, particularly if computer
aided diagnostic techniques are to
play any future role in this area.
Knowledge of the quality of the
image data which is being assessed
will determine the quality of any
outcome whether this is assessed
by human observers or automati-
cally by computer techniques.
Understanding fully what consti-
tutes quality in medical x-ray
images will also help in the
development and implementation
of new imaging technology
including methods for display,
transfer and storage of image data.
20
Page 20
Unification of physical and clinical
requirements for medical X-ray imaging
and its relevance to European industrial
and socio-economic development
(UCP X-IM)
Challenges to be met
So far all attempts to employ the presenta-
tion of clinical details in methodologies for
image quality assessment have generally
proved to be inconclusive. This is particularly
true when taking into account the need to
consider both normal anatomy and abnor-
mal pathological structures that may be pre-
sent on the same image. The main reason
for this situation lies in the fact that assess-
ment of x-ray images by doctors usually
involves a subjective search for an abnor-
mality. In effect each doctor reporting on
an image utilises his own inbuilt clinical
criteria for determining the presence or
absence of such abnormalities. These crite-
ria are programmed into a doctor’s memory
through experience gained either as part of
formal medical training or afterwards whilst
involved in clinical practice throughout his
working lifetime.
Attempts to define image quality scientifically
have, so far, concentrated on defining and
assessing physical parameters associated
with the imaging process itself. Such para-
meters, such as noise, resolution, contrast etc,
concentrate on particular aspects of the “This is not a quality X-ray”
imaging process. However a clinical x-ray
image reproduces a variety of structures of
different shapes and sizes and to a large
extent it is the overall reproduction of these Achievements
which defines image quality. Such a problem
is further complicated by the possible pres- During the past 10 years work has been
ence or absence of abnormal pathological ongoing in Europe, as part of the European
structures, which may be present at the bor- Commission’s Radiation Protection Research
derline of visual detectability. Programme, aimed at developing sets of
quality criteria for common types of diag-
Historically, an accurate and meaningful nostic radiographic images. These include
description of normal and abnormal images of the chest, lumbar spine and breast
anatomical structures that may be repro- which are either frequently occurring or rel-
duced in x-ray images was not considered atively high dose examinations. The criteria
necessary. Those descriptions that did exist specify linguistically and in some detail what
were employed solely as part of the educa- normal anatomical structures should be
tion and training programme for each gen- reproduced in good quality radiographic
eration of radiologists. That such descriptors images of a particular examination. This
might play a role in assessing image quality concept was evaluated during two large
was not considered possible. This situation scale European wide trials and the feasibil-
has prevailed for almost one hundred years. ity of such an approach was demonstrated.
Medecine.EN (Flash) 23/10/02 16:34 Page 21
As part of the 4th Framework Programme cri-
teria were further refined in order to improve
linguistic robustness. Also, the methodology
for implementing such an approach rou-
tinely has been investigated. Measurements
of image quality assessed by means of these
criteria were, for the first time, linked to
other objective assessment methods which
employ visual detection techniques. Those
employed were visual grading (VGA) and
receiver operating characteristic (ROC) analy-
sis. This work also highlighted the role of the
viewer or person undertaking the visual
assessment in the overall scheme. Part of
this work involved a series of trials employ-
ing panels of expert radiologists who “read”
a large number of images produced under
very controlled conditions. This experimen-
tal programme was underpinned by a
detailed theoretical modelling programme
aimed at gaining a much deeper under-
standing of the role and relevance of clini-
cal structure in the imaging process.
The 4th Framework Programme employed
screen-film as the x-ray imaging medium
since this is still the most widely used radi-
ographic method and represents the “gold
standard”. Work in the 5th Framework
Programme is now concentrating on apply-
ing experience gained to date on the assess-
ment of image quality produced by new
totally digital x-ray imaging systems.
PART 2
UCP X-IM Information Column
Title:
Unification of physical and clinical
requirements for medical X-ray
imaging and its relevance to European
industrial and socio-economic
development (UCP X-IM)
Co-ordinator:
B. Michael Moores
Integrated Radiological Services Ltd
The variety of technologies that will be stud- Unit 188, Century Building
ied has already been selected and will provide Brunswick Business Park
a deeper understanding of the role and rele- 102 Tower Street
vance of the two stages in image formation, UK-L3 4BJ LIVERPOOL,
namely x-ray detection and image display. United Kingdom
Tel: +44.151.709.6296
Further attempts to strengthen the rela- Fax: +44.151.709.8772
e-mail: bmm@irs.u-net.com
tionship between the assessment of image
quality by means of quality criteria to that Partners:
employing ROC analysis are at an advanced - Sören Mattsson
stage of development. Similarly methods (Lund University, Sweden)
are now in place to generate mathematically - Werner Panzer
abnormal pathological structures which can (GSF Neuherberg, Germany)
be superimposed on normal x-ray images in - David Dance
order to investigate the relevance of normal (Royal Marsden NHS Trust,
and abnormal anatomy in defining a qual- London, UK)
ity image. A set of physical descriptors for - Gudrun Alm Carlsson
defining imaging performance of new dig- (Linkoping University, Sweden)
- Francis Mer
ital x-ray systems has also been developed.
(Institut Radiophysique Appliquee,
Lausanne, Switzerland)
Over the remaining part of the project both - Egbert Buhr
physical and clinical assessment method- (PTB, Braunschweig, Germany)
ologies will be applied to real systems and - Christoph Hoeschen
the interplay between them will be fully (Otto-von-Guericke University,
investigated. Because of the spread of exper- Magdeburg, Germany)
tise and resources amongst the collaborat-
ing partners, the project is able to tackle this EC Scientific Officer:
wide ranging problem, a situation which has Ernst-H.Schulte
not been possible previously. Similarly Tel: +32.2.295.71.55
Fax:+32.2.295.49.91
because this project is built upon previous rel-
e-mail:
atively long term progress within this area
ernst-hermann.schulte@cec.eu.int
an evolution of concepts and underlying phi-
losophy has been possible. Period:
Nuclear Energy 1998-2002
Status:
Ongoing
21
Medecine.EN (Flash) 23/10/02 16:34
bjectives
Screen film radiography is still the
primary means of X-ray radiography
for many reasons, including the
cost considerations, the radiologists’
attachment to traditions and the
fact that the new detectors are not
yet widely available. The desire to
make the transition to digital
radiography is widely acknowl-
edged, recognising that it should
lead to increased radiology depart-
ment productivity as well as yield-
ing diagnostic/dose benefits
resulting primarily from the higher
radiation collection efficiency and
dynamic range of new detectors.
The dawning new era in detector
technology automatically raises
the issue of its impact on radiation
protection.
This Concerted Action has aimed
at studying the challenges created
by new detectors with emphasis
on radiation protection aspects.
The basic new detector technologies
were to be described and their
strengths and weaknesses identified.
The figures of merit quantifying
such strengths and weaknesses
were to be established. The role
of new materials in detector
technology was to be highlighted
together with the research role
and strategic objectives of
the major industrial manufacturers
and academic institutions.
The attitude of the clinical sector
with respect to new detector
technologies was to be investigated.
22
Page 22
New detector technologies
in x-ray diagnosis and impact
on radiation protection strategies
Challenges to be met
New detector technology was investigated
from a variety of different standpoints rang-
ing from fundamental research to potential
impact on both the manufacturers and users.
The basic detector research review com-
prised a historical section and a survey of
ongoing activities in Europe and worldwide.
The review tried to identify the driving force
behind each group for basic research on a
particular detector technology, e.g. previous
experience and infrastructure or the use of
the same detector type for other purposes.
The fundamental physical processes involved
in detector technology, for signal detection
as well as processing techniques and data
acquisition were also studied.
The applied and industrial considerations
covered two steps in the development
process: Detector to Prototype and
Prototype to Industrial Product. The former
evaluated the mechanisms for implement-
ing basic research outcomes, including pro-
totype development and its interaction with
and possible impact upon the clinical envi-
ronment. The latter concentrated on the
manufacturing and marketing of a new tech-
nology and the mechanisms for introducing
it to routine clinical implementation.
Incident X-ray radiation
Detection substrate (CdZnTe or Si)
Readout substrate (CMOS)
Support structure (PCB or ceramic)
Figure 1 – A schematic detail of the SIMAGE generic technology
Under industrial and strategic considera-
tions the impact of strategic and financial
support for an industrial action was inves-
tigated with respect to implementation of
a new medical technology, in particular, a
new x-ray detector technology. The per-
ception of technological development by
government agencies was evaluated. Also
reviewed was the fulfillment of Euratom
Directives in the field of radiation protec-
tion and the role of EU government agen-
cies in supporting and funding the end user
developments.
Achievements
Most of the research and development activ-
ity in conventional radiography is concen-
trating on the Active Matrix Flat Panel
Imagers based on amorphous silicon TFT
(Thin Film Transistor) arrays. A promising
technology in digital radiography utilizes
the hybrid solid state semiconductors that
are constructed from a flip-chip joined to an
x-ray stopping layer, such as Si, GaAs, CdTe
and CdZnTe with a CMOS (Complementary
Metal-Oxide Semiconductor) readout chip.
SIMAGE, a Finnish company, has developed
such detectors based on Si and CdZnTe.
Metallization
Pixel
metal
contact
Bump
CMOS
metal
contact
Medecine.EN (Flash) 23/10/02 16:34
These detectors have proven to be promis-
ing particularly in small area applications
such as dental imaging or small area real
time imaging.
Figure 2 – SIMAGE CMOS detector image – direct digital X-ray image
Other developments include improved
Charge Coupled Device detectors, or
detectors with a scintillator directly
attached to a CMOS readout chip.
Experience from experimental high-energy
physics has prompted the consideration
of silicon strip detectors and new types of
detectors based on gas amplification in
x-ray imaging applications.
All detector R&D has been directed towards
the creation of a prototype product. A sub-
stantial difference has been identified
between industrial based R&D and univer-
sity R&D. Clear differences have been also
observed as a function of the intended
application. Clinical knowledge suggests
that the link between detector perfor-
mance, as expressed in technical terms,
and clinical diagnostic value is not direct.
A cultural difference between European and
American manufacturers concerning their
approach to new detectors is also apparent.
European manufacturers are aware that the
radiation safety culture embodies the need
to optimise protection so that improved
radiation detection should lead to reduced
doses. On the other hand, the North
American culture would focus on the use of
new detectors to obtain higher signal-to-
noise ratios.
Page 23
PART 2
Information Column
Title:
Marketing considerations are an important New detector technologies in x-ray
stimulus to detector development in certain diagnosis and impact on radiation
cases. Products may well be developed with protection strategies
features that offer no significant clinical
Co-ordinator:
advantage. Dose reduction
Risto Orava
measures may however be Helsinki Institute of Physics
incorporated as a positive fea- CERN Office
ture in the marketing strategy CH-1211 GENEVE 23,
of equipment manufacturers. Switzerland
Tel: +41.22.767.46.96 or
Although not their original +358.9.191.50686
objective, several equipment Fax: +41.22.767.36.00
manufacturers are emphasis- e-mail: Risto.Orava@cern.ch
ing dose awareness as part of
their marketing strategy, espe- Partners:
- Militadis Sarakinos
cially for applications where
(HIP-CERN, Geneva, Switzerland)
dose issues are less critical (e.g. - Willi Kalender
chest radiography). In gen- (Institute of Medical Physics,
eral, market-driven research Erlangen, Germany)
aims at offering the next step - Sven Kulander
in marketable products (e.g. (Uppsala University, Sweden)
the next generation of flat - Hans-Peter Busch
panel detectors). However, (Brüderkrankenhaus, Trier, Germany)
there is also detector-driven research as a - Karl-Friedrich Kamm
spin-off from non-medical applications. (Philips Medical Systems, Hamburg,
Germany)
- Michael Moores
University research is much more diverse.
(IRS, Liverpool, UK)
Fundamental research is the first step which
may or may not lead to marketable innova- EC Scientific Officer:
tion but fundamental research without mar- Diederik Teunen
ketable or applicable prospects is absorbing (Contact: George Neale Kelly)
a lot of resources without any apparent return. Tel: +32.2.295.64.84
Fundamental research should be seen as a Fax:+32.2.295.49.91
long-term investment with a return that is not e-mail: george-neale.kelly@cec.eu.int
quantifiable with respect to specific projects.
However, it is clear that there is an enormous Period:
margin for improvement by encouraging fun- Nuclear Energy 1994-1998
damental research to take account of the cur-
Status:
rent market status. Furthermore, many
Completed
fundamental research institutes do not even
quantify detector performance in terms of
the usual figures of merit used in the medical
imaging community.
New detector technologies are entering the
world of medical x-ray imaging. While the
market is still at an early stage, the poten-
tial seems large. The advantages of new
technology, in terms of image capture as
well as productivity strongly support this
trend. Which new detector technology will
actually dominate the market is presently
unclear. It should be noted that the major-
ity of the earth's population has never had
an x-ray examination; thus the potential
market for new technologies is huge.
23
Medecine.EN (Flash) 23/10/02 16:34
bjectives
Oral radiography is one of the most
frequently used diagnostic tools in
dentistry. Although the estimated
risk from exposure to oral
radiography is relatively small in
comparison with other risks, its
frequent use and rapid technological
evolution necessitate further
research and documentation on
quality assurance. Optimisation of
the whole radiographic procedure
is the best way to ensure the
maximum patient benefit with a
minimum of patient and operator
exposure. In this context it should
be realised that similar diagnostic
information can often be obtained
at considerably lower radiation
doses, depending on the
radiographic procedure and the
technical and physical parameters
involved. The overall aim of the
MINOSQUARE concerted action
was to attempt to validate digital
imaging techniques and “Hi-Tech”
multimodal equipment, to
determine optimisation strategies
and to provide guidelines for
quality assurance of such new
developments.
24
Page 24
Maxillary imaging network on optimisation
strategies and quality assurance of
radiographs in Europe (MINOSQUARE)
Challenges to be met
To determine optimisation strategies and
to provide recommendations for quality
assurance of digital oral imaging and
advanced extra-oral imaging procedures,
a number of issues had to be addressed.
These included the establishment of quality
criteria for digital imaging; optimisation
and standardisation of image transmission;
the validation of advanced diagnostic equip-
ment for extra-oral imaging; the assessment
of radiation doses for advanced oral imag-
ing; and the formulation of recommenda-
tions for optimisation strategies and quality
assurance for advanced oral imaging.
Figure 1 – Digital oral imaging allows for image manipulation, as visualised on
the present panoramic image in a female adolescent.
Achievements
The fact that oral imaging is a rapidly
expanding procedure and exploits changing
technology emphasises the need to take
new initiatives to determine quality criteria
for digital oral imaging. The MINOSQUARE
concerted action has defined the quality
criteria and provided information on the
technical and clinical performance of the
currently available digital imaging tech-
niques. The use of digital oral imaging not
only allows for a reduction in radiation
dose, procedure time and film costs, but also
enables more efficient communication
between practitioners and specialists.
To allow easy and universal access to tele-
dental services, organisational aspects, tech-
nological requirements and user needs have
been identified and test results for imple-
mentation in dentistry have been reported.
Besides digital intra-oral imaging systems,
Hi-Tech multimodal equipment for extra-
oral imaging has also been developed which
permits visualisation of panoramic, lateral,
cross-sectional and other detailed views of
the jaw bones.
The Cranex Tome® multimodal radiation
unit manufactured by Soredex, Orion cor-
poration, Finland was technically and clini-
cally validated and proven to provide an
acceptable image quality at a low radiation
dose level.
Dose measurements were also performed for
a number of other applications, specifically
in relation to digital imaging. The accumu-
lated data obtained were then used to for-
mulate recommendations for optimisation
strategies and quality assurance of digital
and advanced extra-oral imaging. A number
of publications in international peer-
reviewed journals resulted from the work.
Medecine.EN (Flash) 23/10/02 16:34 Page 25

PART 2
MINOSQUARE Information Column

Title:
Maxillary Imaging Network on
Optimisation Strategies and Quality
Assurance of Radiographs in Europe
(MINOSQUARE)

Co-ordinator:
Reinhilde Jacobs
Oral Imaging Center
K.U. Leuven-Catholic University Leuven
Kapucijnenvoer 7
B-3000 LEUVEN,
Belgium
Tel: +32.16.33.24.10
A B Fax: +32.16.33.24.84
Figure 3 – Spiral tomography of the upper jaw bone (Cranex Tome ) ® e-mail:
was validated by comparing dimensional measurements on sectioned reinhilde.jacobs@med.kuleuven.ac.be
cadaver bone (A) with those on cross-sectional radiographic images (B).
Partners:
- Paul Frans Van der Stelt
(Academisch Centrum
Tandheelkunde, Amsterdam,
The recommendations coming from this pro- Partnership The Netherlands)
ject have also been published (Jacobs R., - Jean Loup Coudert
Gijbels F. (2000) Oral Imaging 2000. Edmin, The MINOSQUARE concerted action brought (Université Claude Bernard, Lyons,
Department of Periodontology, KULeuven: together the accumulated expertise of France)
Leuven. ISBN90-804618-2-2). This book researchers, clinicians, technology produc- - Michel Jourlin
reflects the contribution of the different part- ers, manufacturers and distributors involved (Ecole Supérieure de Chimie,
ners reported on the International Workshop in dentomaxillofacial imaging. This European Physique, Electronique de Lyon,
France)
on Oral Imaging 2000, held in January 2000 project formed part of the activities
- Christian Lamorall
in Leuven, Belgium. Cross-fertilisation of performed by an established group of (Lamoral N.V., Brugge, Belgium)
ideas at this event was guaranteed by the researchers, the European Dentomaxillofacial - Matti Linnosaari
attendance of more than 30 participants Imaging Network (EDMIN). The excellent (Orion Corporation Soredex,
including key experts from dentistry, radiol- collaboration enabled the objectives of the Helsinki, Finland)
ogy, biomechanical engineering and medical MINOSQUARE project to be achieved and - Jean-Marc Cannaméla
research and development. The results of resulted in the determination of optimisa- (Medical Universal, Lyons, France)
this successful project should help to promote tion strategies and guidelines for quality
digital imaging and advanced oral imaging assurance of new developments and digital EC Scientific Officer:
modalities in dental practice. techniques in dentomaxillofacial imaging. Diederik Teunen
(Contact: George Neale Kelly)
Tel: +32.2.295.64.84
Fax:+32.2.295.49.91
e-mail:
george-neale.kelly@cec.eu.int

Period:
Nuclear Energy 1994-1998

Status:
Completed

Figure 2 – A Cranex Tome ® multimodal radiation unit (Soredex, Helsinki, Finland)

was technically and clinically validated during the Minosquare concerted action.

25
Medecine.EN (Flash) 23/10/02 16:34
bjectives
The main objective of this project
was to draft new European
Guidelines providing quality
criteria for the relatively high-dose
medical x-ray examinations carried
out on children using fluoroscopy
and computed tomography (CT).
This was to be achieved by
conducting surveys of paediatric
radiology practice throughout
Europe and by developing suitable
practical techniques for monitoring
and comparing both patient doses
and image quality that are
particularly adapted to x-ray
examinations on children.
Child being examined on a CT scanner
26
Page 26
Radiation protection of
the patient in paediatric radiology
Challenges to be met
The population of Europe is exposed to
higher levels of radiation, both individually
and collectively, from medical radiology than
from any other man-made source of ionising
radiation and children are appreciably more
sensitive to the effects of radiation than
adults. Paediatric radiology is consequently
an important area for radiation protection
initiatives to ensure that the doses received
by children are the minimum compatible
with achieving the required clinical objective.
This is particularly true for x-ray imaging
modalities that result in relatively high indi-
vidual doses, such as fluoroscopy and CT.
In previous EC Framework Programmes, the
optimisation of common radiographic exam-
inations on children was studied and culmi-
nated in a publication [European Guidelines
on Quality Criteria for Diagnostic Radiographic
Images in Paediatrics, EUR 16261 June 1996].
These guidelines contained both patient dose
and image quality criteria. The dose criteria
were based on the results of European patient
dose surveys and used the concept of refer-
ence doses designed to trigger an investiga-
tion when doses were abnormally high.
However, problems arose due to the con-
founding influence of large variations in the
size of children, even when restricted to nar-
row age bands. The image quality criteria
were expressed in terms of the required visi-
bility of anatomical details that were essen-
tial for diagnosis. They had been developed
by an expert group of paediatric radiologists
from six European countries - the European
Network of Paediatric Radiologists (ENPR)
but, being purely subjective, there were dif-
ficulties in their consistent interpretation.
The quality criteria concept had also been
applied previously to CT examinations and
developed to the stage of a Working Docu-
ment on Quality Criteria for Computed
Tomography. However, the guidance was
tailored to the needs of adult patients and
no attempt had been made to adapt it to the
different requirements for imaging babies
and children.
In this project medical physicists collabo-
rated with the radiologists of ENPR to
develop more robust patient dose and image
quality criteria that could be applied to chil-
dren of any age (or size) undergoing high-
dose fluoroscopic and CT x-ray examinations,
where radiation protection is paramount.
Achievements
Quality criteria guidelines for fluoroscopic
examinations
A draft document giving European
Guidelines for the Optimisation of
Fluoroscopic Imaging in Paediatrics was pre-
pared by the ENPR. The format is similar to
the earlier European Guidelines for radi-
ographic examinations in paediatrics [EUR
16261] and covers eight types of common
fluoroscopic examination on children. The
quality criteria cover additional aspects of
imaging technique that are specific to fluo-
roscopic examinations such as the volume
and density of contrast medium and the
anatomy to be imaged by fluoroscopy or
spot films. The image criteria describe
anatomical features that should be visible in
the spot film images. The dose criteria are
expressed in three ways - the fluoroscopic
screening time, the number of radiographic
exposures and the total Dose-Area Product
(DAP) for 0, 1, 5, 10, and 15 year old
patients. Dose criteria were derived and the
image quality criteria tested in a survey of
micturating cystourethrography (MCU) -
the most frequent fluoroscopic examination
on children – at 12 European hospitals.
Quality criteria guidelines for CT examinations
ENPR drafted a set of quality criteria cover-
ing nine types of CT examination of partic-
ular importance in paediatrics, similar to
those developed earlier for adult patients
[EUR 16262]. The image criteria were
changed to be specific to the anatomy and
pathology of paediatric patients. The dose
criteria were expressed in the same dose
quantities as for adults, i.e. the weighted CT
dose index (CTDIW) and the dose-length
Medecine.EN (Flash) 23/10/02 16:34
Existing European Guidelines for Radiation Protection of Patients in
Diagnostic Radiology
product (DLP) but with a smaller phantom
for children. Initial dose criteria were estab-
lished for babies, 5 and 10 year old children
by sending a questionnaire to 50 European
hospitals asking for details of their exami-
nation protocols for the nine types of CT
examination.
Patient dose criteria
A method was developed for normalising
dose measurements on patients of any size
to the nearest of a set of standard-sized
patients representing children of 0, 1, 5, 10
and 15 years old. Factors were derived for
normalising both entrance surface dose
(ESD) measurements for individual radi-
ographs and DAP measurements for com-
plete fluoroscopic examinations. ESD
measurements from the earlier European
paediatric dose survey reported in EUR
16261 and DAP measurements from the
MCU survey in this project were normalised
in this way. The distributions of the size-
normalised doses were used to derive new
age-specific dose criteria for common radi-
ographs and MCU examinations, based on
the rounded third quartile values.
Image quality criteria
To improve the scientific basis for the image
quality criteria, two studies were made of the
correlation between radiologists’ subjective
assessments of image quality and more
objective physical measures using digitised
images of test objects. The first study con-
centrated on chest radiographs, where the
signal to noise ratio (SNR) in the test object
images showed the best correlation with the
ENPR radiologists subjective assessment of
a series of chest radiographs taken on five
pigs simulating children of 1-7 years old.
Page 27
PART 2
Information Column
Title:
Radiation protection
of the patient in paediatric radiology
Co-ordinator:
Barry F. Wall
NRPB-National Radiological
Protection Board
Chilton, Didcot
UK- OX11 ORQ OXON,
United Kingdom
Tel: +44.1235.822645
Fax: +44.1235.833891
e-mail: barry.wall@nrpb.org
Partners:
- Karl Schneider
(Dr Von Haunersches Kinderspital,
Munich, Germany)
However, the correlation was not sufficient - ENPR-The European Network
to enable a more precise optimisation of of Paediatric Radiologists
imaging technique. In the second study, on - A. Servomaa
paediatric fluoroscopy images, the SNR of (STUK, Helsinki, Finland)
iodine and barium contrast details in phan- - M. Fitzgerald
toms simulating 0, 3, 10 and 15 year old (Radiological Protection Centre,
patients was measured. The SNR and the St George’s Hospital, London, UK)
dose to the phantoms were estimated and
optimal exposure conditions derived as those EC Scientific Officer:
that minimise the dose while maintaining a Diederik Teunen
(Contact: Christian Desaintes)
constant SNR. Contrary to the advice in ear-
Tel: +32.2.295.82.73
lier European guidelines, the optimum expo-
Fax:+32.2.295.49.91
sure conditions for imaging low-moderate e-mail: christian.desaintes@cec.eu.int
contrast details were found to be a rela-
tively low tube voltage of 50-60 kV, high Period:
additional filtration of 0.25 mm Cu and the Nuclear Energy 1994-1998
use of a fibre interspaced antiscatter grid
(except for babies). Recommendations to Status:
this effect were included in the draft qual- Completed
ity criteria guidelines for paediatric fluoro-
scopic examinations.
Partnership
This collaboration between European radi-
ologists and medical physicists has improved
the scientific basis for guidance on good
paediatric imaging techniques and has pro-
vided a practical methodology for estab-
lishing image quality and patient dose criteria
against which paediatric imaging depart-
ments throughout Europe can compare their
performance. The two draft quality criteria
documents provide the framework within
which complete European Guidelines can be
developed. They could be made available to
the radiology community in their present
form as Draft Working Documents or be devel-
oped into more complete European
Guidelines in future EC Framework
Programmes, following more extensive tri-
als of the image criteria and patient dose sur-
veys for a wider range of examinations.
27
Medecine.EN (Flash) 23/10/02 16:34
bjectives
In paediatric radiology it is difficult
to find a statistically significant
correlation between image quality
and technical parameters because
of the large variation in size of
patients and thus in image and
dose response to a diagnostic
exposure. However, establishing
such a correlation becomes very
important because of the higher
sensitivity to radiation of young
patients. The overall objective of
this project was twofold. Firstly to
investigate the link between the
information content of a radiological
image and the associated imaging
procedures. Secondly to develop
optimisation strategies for
paediatric patients by deriving
from adult data a specific
technical-clinical basis for the
quality criteria concepts.
28
Page 28
Optimisation of radiation protection
of the patient in paediatric radiology
(ORPED)
Challenges to be met
The main part of the work was to collect
and analyse data on image quality for chest
and lumbar spine examinations in paediatric
radiology. In particular the relationship
between image quality, dose and technical
parameters was to be investigated. Other
problems related to the criteria used for
rejecting films and the impact of image detail
detectability on the diagnostic information
resulting from the exposure. The scientific
background for data analysis applicable to
the domain of paediatric radiology was to be
ascertained. Finally it was proposed to inves-
tigate the links between patient character-
istics and optimised technical parameters
(linked to a pre-defined level of image qual-
ity). From this it was hoped to define the
most suitable technique to be used to opti-
mise paediatric radiographic procedures.
Achievements
In order to establish a sound scientific back-
ground for data analysis, dose and image
quality data from patients in paediatric radi-
ology were collected. Data related to imag-
ing systems of either the conventional or the
digital type, the latter being located in an
emergency room. Together with relevant
dosimetric results including 278 thermolu-
minescent dosimeter (TLD) measurements,
they were entered into a database for mul-
tivariate statistical analysis.
As an input to a film reject analysis 1,428
chest and 1,181 full spine examinations from
three the last years were studied resulting
in an average rejection rate of 5.5 %. A
detailed analysis of the cause of rejection was
performed leading to the establishment of
a methodological framework for the defin-
ition of discrimination factors and their eval-
uation as classification criteria. Arranging
observations into groups according to their
characteristics was generally achieved by
multivariate statistical analysis.
For assessing the importance of the detec-
tion of image details and their impact on
diagnostic information, a further screening
of rejected chest radiographs was carried
out both for conventional and digital
images (36 and 37 respectively). After a
first analysis a modified list of criteria was
proposed (for example 7 criteria for the
chest). Then rejected images were scored
according to the number of criteria that
were not met, ranging from 10% (trachea)
to 40% (thorax). In general, criteria were
well met except for positioning (full inspi-
ration), some applied techniques (collima-
tion) or in identifying hidden pathologies.
As a result of this work a final list of image
criteria was selected.
Investigating the links between image qual-
ity, dose and technical parameters required
the use of a dedicated test object. A 2-D
motor controlled phantom was designed
and assembled with the preliminary goal of
emulating patient movement in clinical sit-
uations. In the first place, this test object
(PAMOSITO) was used for the selection of
the most suitable screen-film system for
chest radiography and for the optimisation
of radiographic technique.
Thereafter the test object was used for
assessing the influence on image quality of
patient motion in digital fluoroscopy or
cine-fluorography systems. Degradation of
image contrast caused by motion was eval-
uated (depending on frame rate and pulse
width). An increase of the object’s speed up
to 40 mm/s was found to lead to a signifi-
cant loss of image contrast (about 20%)
compared with a motionless object. In this
way criteria could be developed for selec-
tion of technical parameters and scientific
data were gathered to design a second pro-
totype to study cardiac-patient movements,
a situation frequently met within the field
of paediatric radiology.
Medecine.EN (Flash) 23/10/02 16:34
PAtient MOvement Simulation Test Object (PAMOSITO) was constructed with
modular parts to use different mobile test objects and static structures. The
system allows programming different cycles of movement along two axes (x
and z). PAMOSITO features two step motors for the z axis with a 50 mm
range to simulate the patient respiratory movement and small displacements
in X-ray oblique projections. The test object can be moved along the x axis by
means of a linear motor with a 145 mm range.
Analysis of chest examination data was per-
formed to optimise technical parameters
taking into account patient characteristics.
Three clusters of image criteria showed par-
ticular links with image quality. These were
those related to physical parameters (mAs, kV
etc.), those linked to the potential pathology
(signs of pathology) and those related to an
increase of noise of the image information
content (positioning, bad symmetry etc.).
By according a ‘weight’ to each criterion a
tool was created to discriminate an accept-
able from a “potentially” rejected image.
Results obtained in this way coincide rather
well with experimental data from one of
the participating laboratories.
The definition and implementation of opti-
misation strategies in paediatric radiology
was based on the study of 104 chest exam-
ination films (PA projection). It was found
that dose correlates with patient size indices
(diameter or thickness) so a first attempt was
Page 29
PART 2
ORPED Information Column
Title:
Optimisation of radiation protection
of the patient in paediatric radiology
(ORPED)
Co-ordinator:
Carlo Maccia
CAATS-Centre d'Assurance de qualité
des Applications Technologiques
dans le domaine de la Santé
Boulevard Maréchal Joffre 43
F-92340 BOURG-LA-REINE,
France
Tel: +33.1.41.87.66.70
Fax: +33.1.46.61.56.00
e-mail: mcarlo@compuserve.com
Partners:
- Eliseo Vañó
(Universidad Complutense de Madrid,
Spain)
- Gerhard Stoeffler
(Karl-Franzens-Universität Graz,
Austria)
EC Scientific Officer:
Diederik Teunen
(Contact: George Neale Kelly)
Tel: +32.2.295.64.84
Fax:+32.2.295.49.91
e-mail: george-neale.kelly@cec.eu.int
Period:
Nuclear Energy 1994-1998
made to define normalised dose. The values Status:
obtained were 0.099 mGy for PA chest (4-6 Completed
years old patients, 11 cm thickness) and
3.79 mGy for full spine (10-14 years, 17 cm
thickness) respectively.
In order to look for the best technique to be
used in digital and conventional radiogra-
phy, links between exposure technique and
image quality (using Leeds test object TOR,
CDR and an acrylic phantom) were studied.
The first finding was that simulated chest
examinations in digital and conventional
radiography (speed 400) have a compara-
ble spatial resolution value of about 3.55
lines/mm. The other most relevant finding
was that image quality could be altered by
increasing the tube filtration (for example:
from 0 to 1 mm Al + 0.2 mm Cu), which
reduces the detectability of low contrast
details; and by using a grid, which improves
the image contrast (increased number of
contrast details) in conventional radiogra-
phy in particular.
29
Medecine.EN (Flash) 23/10/02 16:34
bjectives
Knowledge of patient doses serves
to stimulate the awareness of the
medical staff with regard to
radiation protection of the patient.
Patient doses vary due to differences
in the technical specifications of
the equipment used and the
diagnostic procedures applied. For
these reasons expertise on radiation
dosimetry and appropriate technical
facilities should be available to all
those using diagnostic x-ray
installations. Although dose
measurements should preferably
be made in house, it is recognised
that dosimetric resources, such as
well-trained specialists, high quality
equipment and calibration facilities
are not always available locally.
However, appropriate assistance
and dosimetric services can often
be obtained from third parties.
The main objective of this project
was to ensure that dose measure-
ments made by laboratories
anywhere in the Member States of
the European Union are based on
the same principles and methods.
To ensure compatibility of the data
collected, it is essential that opera-
tional protocols are strictly followed
and that calibrations are traceable
to appropriate primary standards.
Therefore, this concerted action
aimed at developing a document
providing recommendations for
patient dosimetry in diagnostic
radiology using thermoluminescent
dosimeters (TLDs).
30
Page 30
Criteria and protocols for central laboratories
providing dosimetric services for
quality control in diagnostic radiology
(DIAGNOSDOS)
Challenges to be met
Two types of users of the document were
foreseen. The first is persons in laboratories
specialising in dosimetry using TLDs, the
second is staff of radiology departments.
The former may not necessarily be aware of
the concepts of patient dosimetry in diag-
nostic radiology. The latter may lack exper-
tise on dosimetry using TLDs. Therefore,
extensive background information had to be
provided in the document on basic dosi-
metric concepts and measuring techniques
as well as on organisational and adminis-
trative aspects.
Achievements
The recommendations were prepared by a
drafting group set up under the project and
reviewed by advisory groups who provided
comments and additions. The final draft
was then circulated among a larger group
of thirteen experts, some of whom repre-
sented professional societies, for further
comments. The final draft was also discussed
during a workshop.
The final document resulting from the dis-
cussions covers both the conceptual and
practical aspects of patient dosimetry in
general diagnostic radiology, mammography
and computed tomography. [Recommenda-
tions for Patient Dosimetry in Diagnostic
Radiology Using TLD, Nuclear Science and
1 Figure 1 – Calculated relative
0.95
0.9
C80 Free-in-air
Relative response
C60 Free-in-air
0.85 Mo/Mo PMMA-phantom
Mo/Mo Free-in-air
0.8
0.75
0.7
0.65
0.6
0 15 30 45 60 75
Angle of radiation incidence (degree)
Technology, Topical report, EUR 19604 EN,
2000]. The basic dosimetric concepts and
reference quantities, such as entrance sur-
face dose when the patient is present
(ESDpp), entrance surface dose when the
patient is absent (ESDpa), dose area product
(DAP) and computed tomography dose
index (CTDI) are presented.
Dose measurements can be made both on
patients and on phantoms. For such mea-
surements ionisation chambers and TLDs are
widely used. This document concentrates on
the application of TLDs, for which not only
the principles but also many practical aspects,
including pitfalls, are described. The response
of a TLD depends on a large number of para-
meters, some of which are inherent in the
characteristics and treatment of the ther-
moluminescent phosphor, others depend on
the detector and reader design. Those that
are inherent are properties such as the lin-
earity of response with dose, the sensitivity
to exposure to light, the intrinsic TL efficiency
depending on photon energy (LET depen-
dence) and the fading characteristics of the
phosphor. Some of these can be modified by
procedures including the heat treatment of
the TL detectors (annealing). The detector
design determines the absorption coefficients
of the detector material, the self shielding
caused by the detector mass, the absorption
and scattering caused by the wrapping of
the detector, and the variation of response
with angle of incidence of the radiation.
response of a 3.2 mm x 3.2 mm x
0.9 mm LiF TLD-700 detector as a
function of orientation of the
detector during irradiation in
Mo-filtered 28 kV (Mo anode),
80 kV (C60; HVL = 2.76 mm Al)
and 120 kV (C80; HVL = 6.31 mm
Al) X-ray beams. 90 degree angle
indicates a radiation incidence
perpendicular to the largest sur-
face of the detector. For the 28 kV
X-rays data are presented both for
irradiation free-in-air and for
90 irradiation at a depth of 5 mm in
a PMMA phantom.
Medecine.EN (Flash) 23/10/02 16:34 Page 31

PART 2
DIAGNOSDOS Information Column

The readout conditions and the stability of
reader parameters are a function of reader
design and use. Performance requirements
and operational details are discussed in
detail. As an example a schematic presen-
tation of dosimetry procedures is given.
Title:
Criteria and protocols for central
laboratories providing dosimetric
services for quality control in
diagnostic radiology (DIAGNOSDOS)

Table 1. Schematic presentation of procedures involved in a dose measurement for Co-ordinator:

an examination in diagnostic radiology. Johannes Zoetelief
TNO Centre for Radiological
Detectors used in hospital Detectors kept at issuing laboratory Protection and Dosimetry
Action presently, Interfaculty Reactor Institute
Field mea- Background Fading 1 Zero-dose Fading 2 Calibration Delft University of Technology
surements Mekelweg 15
NL-2629 JB, DELFT,
Preparation + + + + + + The Netherlands
Storage at laboratory + + + + + + Tel: +31.15.278.89.87
Test-dose irradiation 2.5 mGy 2.5 mGy Fax: +31.15.278.90.11
e-mail: zoetelief@iri.tudelft.nl
Storage at laboratory + + +
Dispatch to hospital + + + + Partners:
Irradiation hospital/lab. X mGy 2.5 mGy - Hans W. Julius
Dispatch to laboratory + + + (TNO Centre for Radiological
Protection and Dosimetry, Delft,
Storage at laboratory + + + + + +
The Netherlands)
Fresh annealing + - Per Christensen
Pre-read annealing + + + + + + (Risø, Roskilde, Denmark)
Readout + + + + + + - Barry F. Wall
(NRPB, Didcot, UK)
- Michael Moores
(IRS, Liverpool, UK)
The procedures for irradiating the TLDs and Evaluation, interpretation and reporting
- Wolfram Leitz
questionnaires to obtain information about of results are presented. For assessment (SSI, Stockholm, Sweden)
the irradiation conditions are presented. of ESDpp in general diagnostic radiology - Werner Panzer
Detailed instructions to the staff of an x-ray it is necessary to process the TLDs fol- (GSF Neuherberg, Germany)
unit are also provided. As an example, the lowing the standard procedures described
procedure for assessment of ESDpp in gen- above and express the results in terms of EC Scientific Officer:
eral diagnostic radiology is given. ESDpp, calculating the mean value of ESDpp Diederik Teunen
for each projection separately. It is then (Contact: Ernst-H.Schulte)
The first requirement is to keep the control possible to compare the mean value of Tel: +32.2.295.71.55
dosimeters under the same conditions as ESDpp with appropriate diagnostic refer- Fax:+32.2.295.49.91
the dosimeters used for patients, except ence levels. Finally a written report can e-mail:
ernst-hermann.schulte@cec.eu.int
during the irradiation. To obtain reason- be prepared, stating the individual results
able statistical accuracy it is necessary to and the mean value of ESDpp for specific
Period:
select at least 10 adult patients. For each projections and the comparison of the Nuclear Energy 1994-1998
patient and for a relevant projection first mean ESDpp with previous data, if available.
measure the weight of the patient. Then In addition, follow-up actions can be pro- Status:
position one TLD unit on the patient’s skin, posed. Completed
as close as possible to where the centre of Finally some general aspects of quality
the x-ray field enters the patient and x-ray assurance are given.
the patient normally. Remove the TLD unit A future option might be to evaluate the
and keep it away from radiation. Finally recommendations through a demonstra-
complete a questionnaire recording all the tion project involving various partners.
information. When all TLD units have been
used, return them together with the com-
pleted questionnaires to the issuing labo- PARTNERSHIP
ratory. In the questionnaires only those
parameters are listed which are necessary The advantage of the partnership was
to assess and interpret the measured doses. that the recommendations are more gen-
A more detailed questionnaire would be eral than could be obtained at the
needed to find out the reasons for high national level. Thus harmonisation on a
patient doses, or for the purpose of trou- European level is fostered. In addition,
ble shooting; this is given in Appendix I of the partnership provided easy access to
the document. existing national recommendations.

31
Medecine_cover.EN (Flash) 23/10/02 16:41 Page 1

E u r o p e a n C o m m i s s i o n

KI-NA-19-957-EN-C
Community Research

Optimisation of
Protection in
the Medical Uses
of Radiation

OFFICE FOR OFFICIAL PUBLICATIONS

OF THE EUROPEAN COMMUNITIES EURATOM
L-2985 Luxembourg

EUR 19793