Radiation Protection Dosimetry

QUALITY CONTROL AND RADIATION PROTECTION

OF THE PATIENT IN DIAGNOSTIC RADIOLOGY

AND NUCLEAR MEDICINE

Proceedings Editors: G. Contento

. Wall . Schibilla D. Teunen

CON F 930967 ISBN 1 870965 37 X EUR 15257 EN RADIATION PROTECTION DOSIMETRY Published by Nuclear Technology Publishing Vol. 57, Nos. 14,1995

QUALITY CONTROL AND RADIATION PROTECTION

OF THE PATIENT IN DIAGNOSTIC RADIOLOGY

AND NUCLEAR MEDICINE

Proceedings of a Workshop held in Grado, Italy

September 29 to October 1 1993

Co-organised by: Commission of the European Communities

and Unita' Sanitaria Locale n. 7, Udine

Co-sponsored by: The World Health Organisation

Proceedings Editors G. Contento

. Wall . Schibilla D. Teunen

CONF 930967 ISBN 1 870965 37 X EUR 15257 EN RADIATION PROTECTION DOSIMETRY Vol. 57 Nos. 1 - 4 , 1995 Published by Nuclear Technology Publishing

DATA ANALYSIS IN QUALITY CONTROL AND RADIATION PROTECTION OF THE PATIENT IN DIAGNOSTIC RADIOLOGY

AND NUCLEAR MEDICINE.

Grado, Italy, September 29 to October 1 1993

Co-organised by the Commission of the European Communities and the Unita' Sanitaria Locale n. 7, Udine, and co-sponsored by the World Health Organization.

British Library Cataloguing in Publication Data

A catalogue record of this book is available at the British Library

LEGAL NOTICE Neither the Commission of the European Communities nor any person acting on behalf of the Commission is responsible for the use which may be made of the following information.

ECSC-EEC-EAEC Brussels-Luxembourg (1995) Publication No EUR 15257 EN of the Commission of the European Communities, Dissemination of Scientific and Technical Knowledge Unit Directorate-General Telecommunications Information Industries and Exploitation Research, Luxembourg

Radiation Protection Dosimetry ISSN 01448420

EditorinChief: Dr J.A. Dennis, UK

Executive Editor. Mr E.P. Goldfinch, UK

Staff Editor: Mrs M.E. Calcraft, UK

Editorial Board Members Dr R.M. Alexakhin, Russia Prof. Dr K. Becker, Germany Dr M.A. Bender, USA Dr L. BtterJensen, Denmark Dr G. Busuoli, Italy Mr M.W. Carter, Australia Dr M.W. Charles, UK Mr G. Cowper, Canada Dr Li Deping, Peoples Republic of China Prof. Dr B. Drschel, Germany Dr K. Duftschmid, Austria Dr V. I. Fominych, Russia Miss F.A. Fry, UK Mr J.A.B. Gibson, UK Mr R.V. Griffith, USA Dr . Harrison, UK Mr J.R. Harvey, UK Dr H. Ing, Canada Dr . Irlweck, Austria Prof. Dr W. Jacobi, Germany Dr R.L. Kathren, USA

Consultant Editors: Dr G. Dietze, Germany Prof. Y. Horowitz, Israel Dr J. Rundo, USA

Dr E. Kunz, Czech Republic Dr A.R. Lakshmanan, India Dr D.C. Lloyd, UK Mr . O'Brien, USA Dr H.F. Macdonald, UK Mr T.O. Marshall, UK Dr A. Moghissi, USA Dr M. Moscovitch, USA Prof. V. Nishivvaki, Japan Mr E.A. Piesch, Germany Dr G. Portal, France Dr A.S. Pradhan, India Dr D. Ramsden, UK Prof. Dr A. Scharmann, Germany Mr J.A. Selby, USA Dr F. Spurny, Czech Republic Dr R.H. Thomas, USA Mr I.M.G. Thompson, UK Dr L. Tommasino, Italy Mr J.W.N. Tuyn, Switzerland Dr M.E. Wrenn, USA

Published by Nuclear Technology Publishing, P.O. Box 7, Ashford, Kent, TN23 1YW, England.

Advertising office: Mrs L. Richmond Subscription office: Mrs M.L. Mears

Subscription rates: 1994: Vols. 51 5 6 inclusive, UK 510.00 p.a.; outside UK US$ 1 150.00 p.a. 1995: Vols. 5762 inclusive, UK 550.00 p.a.; outside UK US$ 1190.00 p.a.

Orders and remittance should be sent to:

Subscription Department, Nuclear Technology Publishing, P.O. Box No 7, Ashford, Kent, TN23 1YW, England Telephone (01233) 641683 Facsimile (01233) 610021

COPYRIGHT 1995 Nuclear Technology Publishing Legal disclaimer: The publisher, the editors and the Editorial Board accept no responsibility for the content of the papers, the use which may be made of the information or the views expressed by the authors.

Typeset by Photographies, 184 High Street, Honiton, Devon, England Printed by Geerings of Ashford Ltd., Cobbs Wood House, Chart Road, Ashford, Kent, England

EBRPDJ N12

Radiation Protection dosimetry INSTRUCTIONS TO AUTHORS

SCOPE: The scope of this journal covers all aspects of personal and environmental dosimetry and monitoring for ionising and non-ionising radiations, including biological aspects, physical concepts, external personal dosimetry and monitoring, environmental and workplace monitoring and dosimetry related to protection of patients. Animal experiments and ecological sample measurements are not included. Scientific or Technical Papers should be full papers of a theoretical or practical nature with comprehensive descriptions of the work covered. Scientific or Technical Notes should be brief, covering not more than 4 printed pages (one page contains about 800 words or equivalent in figures) and are likely to cover work in development or topics of lesser significance than filli papers. Letters to the Editor should be written as letters with the authors' names and addresses at the end and should be marked 'For Publication'. LANGUAGE: All contributors should be in English. Spelling should be in accordance with the Concise Oxford Dictionary. However please use dosemeter rather than dosimeter, for consistency within the journal. Authors whose mother tongue is not English are requested to ask someone with a good command of English to review their contribution before submission TITLES should be brief and as informative as possible. A short title of not more than 50 characters for a running head should be supplied. AUTHORS' names and addresses (with full postal address) should appear immediately below the title ABSTRACTS containing up to ISO words should be provided on a separate page, headed by the title and authors' names. SCRIPTS must be typewritten and double spaced. One copy must be directly typed and three additional (photocopies) should be provided for refereeing purposes to minimise the time required for refereeing. Headings should be given to main sections and sub-sections which should not be numbered. The title page should contain just the title, authors' names and addresses and a short running title. Manuscripts should be written in the third person and not the first. If your manuscript is prepared using a computer or word processor it would be helpful if you could also send a copy of the computer disc (please specify software). FIGURES AND TABLES should not be inserted in the pages of manuscript but should be supplied on separate sheets. One high quality set of illustrations and figures, suitable for direct reproduction, e.g. black ink or good quality black and white prints of line drawings and graphs, should be provided with original typed manuscript. These should be approximately twice the final printed size (full page printed area = 19cm 15cm). The lettering should be of such a size that the letters and symbols will remain legible after reduction to fit the printed area available. Tables should be typed. Tables should be lightly lined in pencil. All figures and tables should be numbered, using Arabic numerals, on the reverse side of each copy. Numbered captions or titles should be typed on a separate sheet. Figures and tables should be kept to the minimum consistent with clear presentation of the work reported. Half-tone photographs should only be included if absolutely necessary. Figures generated by computer graphics are generally NOT suitable for direct reproduction. Photocopies of all figures and tables should accompany each copy of the manuscript for refereeing purposes. Colour figures can be reproduced at cost. UNITS, SYMBOLS AND EQUATIONS: SI units should be used throughout but other established units may be included in brackets (Note that cGy is not acceptable). Any Greek letters or special symbols used in the text should be identified in the margin on each occasion they are used. Isotope mass numbers should appear at the upper left of the element symbol e.g. "Sr. Equations should be fully typed. FOOTNOTES should only be included if absolutely necessary. They should be typed on a separate sheet and the author should give a clear indication in the text by inserting (see footnote) so that they may appear on the correct page. ABBREVIATIONS which are not in common usage should be defined when they first appear in die text. REFERENCES should be indicated in the text by superior numbers in parenthesis and the filli reference should be given in a list at the end of the paper in the following form, in the order in which they appear in the text:-1. Crase, K.W. and Gammage, R.B. Improvements in the Use ofCeramic BeO in TLD, Health Phys. 29(5) 739-746 (1975). 2. Clarke, R.H. and Webb G.A.M. Methods for Estimating Population Detriment and their Application in Setting Environmental Discharge Limits. Proceedings of Symposium - Biological Implications of Radionuclides Released from Nuclear Industries, Vienna, March 1979. IAEA-SM-237/6, 149-154 (1980) 3. Aird, E.G. A.A. An Introduction to Medical Physics. William Heineman Medical Books Ltd (London). ISBN 0 433 003502. (1983) 4. Duftschmid, K.E. TLD Personnel Monitoring Systems - The Present Situation. Radial. Prot. Dosim. 2(1) 2-12 (1982). All the authors' names and initials (unless there are more than 10 authors), the title of the paper, the abbreviated title of the journal, volume number, page numbers and year should be given. Abbreviated journal titles should be in accordance with the current World List of Scientific Periodicals. If all of this information is not available the reference should not be cited. PROOFS will be sent to any nominated author for final proof reading and must be returned within 3 days of receipt using the addressed label which will be provided. Type-setting or printer's errors should be marked in red. Any other changes should be marked in green but if they are significant they may be charged to the authors. Authors' changes marked in red may not be accepted. The Editor reserves the right to make editorial corrections to manuscripts. An order form for additional reprints will accompany proofs. SUBMISSION: All manuscripts (original and three copies) and correspondence should be addressed to Mr E.P. Goldfinch, Executive Editor, Nuclear Technology Publishing, P.O. Box No 7, Ashford, Kent TN23 1YW, England. It is essential that they are accompanied by six fully addressed adhesive labels addressed to the author nominated to receive proofs and correspondence. These will be used for acknowledgement of receipt of the manuscript, notification of acceptance, return of proofs to authors and supply of reprints. Papers will be considered only on the understanding that they are not currently being submitted to other journals. The Publishers, The Editor-in-Chief and the Editorial Board do not accept responsibility for the technical content, the use of that content or the views expressed by authors. CORRESPONDENCE: Please ensure that you provide telephone, FAX and E-mail numbers if available. Please quote the manuscript number in any correspondence once receipt of your manuscript has been acknowledged. COMPUTER MANUSCRIPTS: If your manuscript is prepared using a computer or word processor, publication may be quicker if you submit a copy of the disc with the manuscript copies. The following programmes can be readily accommodated: - Multimate, Wordstar, MS Word, Word Perfect, Displaywrite, ASCII files and IBM MS DOS Pro Dos Files. COPYRIGHT: Authors submitting manuscripts do so on the understanding that if accepted for publication, copyright of the article shall be assigned to Nuclear Technology Publishing unless other specific arrangements are made. GENERAL: In order to ensure rapid publication it is most important that all of the above instructions are complied with in full. Failure to comply may result in considerable delay in publication or the return of manuscripts to the author. In case of difficulty with illustrations and figures please consult the photo-reprographic section of your establishment. If illustrations of a quality high enough for direct off-set photographic reproduction cannot be supplied they may be redrawn by the publishers at the request of authors if all relevant details are provided. A charge will be made if requirements are extensive.

1NST-J-RPD 4

DATA ANALYSIS IN QUALITY CONTROL AND RADIATION PROTECTION OF THE PATENT IN DIAGNOSTIC RADIOLOGY

AND NUCLEAR MEDICINE.

PROGRAMME COMMITTEE

L. Arranz, Madrid (E) and CGC Radiation Protection D.P. Pretschner, University Hildesheim (D) H. Bergmann, University Wien (A) L. Dalla Palma, University Trieste (I) and EAR P. Elsakkers, University Leiden (NL) and ISRRT G. Hanson, WHO, Geneva J. Liniecki, University Lodz (PL) and ICRP C. Maccia, CAATS, Cachan (F) R. Padovani, USL N.7, Udine (I)

W. Puschert, IEC, Hamburg (D) F.E. Stieve, GSF, Neuherberg (D) O.H. Suleiman, CDRH, Rockville MD (USA) D. Teunen, CEC, Luxembourg (L) J. Valentin, SSI, Stockholm (S) A. Wambersie, UCL Brussels () and ICRU

SCffiNTIFIC SECRETARIAT

G. Contento, USL N.7, Udine (I) H. Schibilla, CEC, Brussels ()

SESSION CHAIRPERSONS

L. Dalla Palma G. Drexler G. Hanson W. Leitz C. Maccia B.M. Moores

R. Padovani . Petoussi D.P. Pretschner H. Schibilla F.E. Stieve A. Talbot

D. Teunen J. Valentin E. Vano Carruana B.F. Wall

REFEREES

M.G. Alm Carlsson . Bauer . Bergmann .. Busch M. Chevalier G. Contento F. Corlob L. Dalla Palma D.R. Dance G. Drexler S. Ebdon-Jackson P. Elsakkers A. Ferro de Carvalho M. Fitzgerald E. Guibelalde O. Hjardemaal K.A. Jessen R. Klausz

H.M. Kramer W. Leitz S. Levialdi C. Maccia M.R. Malisan J.F. Malone S. Mattsson B.M. Moores P. Moran V. Neofotistou J.W. Oestmann R. Padovani M. Paganini-Fioratti W. Panzer N. Petoussi D.P. Pretschner M.L. Ramsdale V. Roberto

P.J. Roberts H. Schibilla T. Schmidt K. Schneider A. Servomaa P.C. Shrimpton F.E. Stieve O.H. Suleiman A. Talbot D. Teunen .A.O. Thijssen G. Tosi J. Valentin E. Vano Carruana B.F. Wall A. Wambersie D. White J. Zoetelief

RADIATION PROTECTION DOSIMETRY Previous Proceedings published by Nuclear Technology Publishing on behalf of the Commission for the European Communities

EXOELECTRON EMISSION AND ITS APPLICATIONS 166 pp, SKIN DOSIMETRY - RADIOLOGICAL PROTECTION ASPECTS OF Softback, Proceedings of the VHth International Symposium, Strasbourg, SKIN IRRADIATION (ISBN 1 870965 12 4) 212 pp, Hardback, March 1983, Price 35.00 Proceedings of a Workshop, Dublin, May 1991, Price 60.00

INDOOR EXPOSURE TO NATURAL RADIATION AND DOSIMETRY IN DIAGNOSTIC RADIOLOGY (ISBN 1 870965 II 6) ASSOCIATED RISK ASSESSMENT 440 pp, Softback, Proceedings of 316 pp, Hardback, Proceedings of a Seminar, Luxembourg, March 1991, an International Seminar, Anacapri, October 1983, Price 70.00 Price 80.00

MICRODOSIMETRIC COUNTERS IN RADIATION PROTECTION AGE DEPENDENT FACTORS IN THE BIOKINETICS AND 120 pp. Softback, Proceedings of a Workshop, Hamburg/Saar, May 1984, DOSIMETRY OF RADIONUCLIDES (ISBN 1 870965 15 9) 254 pp,

Price 25.00 Hardback, Proceedings of a Workshop, Schloss Elmau, November 1991, Price 60.00

RADIATION PROTECTION QUANTITIES FOR EXTERNAL EXPOSURE 166 pp, Softback, Proceedings of a Seminar, Braunschweig, GUIDEBOOK FOR THE TREATMENT OF ACCIDENTAL INTERNAL March 1985, Price 35.00 CONTAMINATION OF WORKERS (ISBN 1 870965 22 1) 50 pp,

Softback, A Joint Publication for the CEC and the USDOE, MICRODOSIMETRY 400 pp, Softback, Proceedings of the Ninth Price 20.00 Symposium on MicrodosimeUy, Toulouse, May 1985, Price 70.00

NEUTRON DOSIMETRY (ISBN 1 870965 16 7) 486 pp, Hardback, DOSIMETRY OF BETA PARTICLES AND LOW ENERGY X RAYS Proceedings of the Seventh Symposium on Neutron Dosimetry, Berlin, 134 pp, Softback, Proceedings of a Workshop, Saclay, October 1985, October 1991, Price 80.00

Price 30.00 THE NATURAL RADIATION ENVIRONMENT (ISBN 1 870965 14 0)

ENVIRONMENTAL AND HUMAN RISKS OF TRITIUM 192 pp, 800 pp, Hardback, Proceedings of the Fifth International Symposium, Softback, Proceedings of a Workshop, Karlsruhe, February 1986, Salzburg, September 1991, Price 120.00

Price 40.00 RADIATION EXPOSURE OF CIVTL AIRCREW (ISBN 1 870965 13 2)

ETCHED TRACK NEUTRON DOSIMETRY 130 pp, Softback, 140 pp, Hardback, Proceedings of a Workshop, Luxembourg, June 1991, Proceedings of a Workshop, Harwell, May 1987, Price 25.00 Price 30.00

ACCIDENTAL URBAN CONTAMINATION 192 pp, Softback, TEST PHANTOMS AND OPTIMISATION IN DIAGNOSTIC Proceedings of a Workshop, Roskilde, June 1987, Price 40.00 RADIOLOGY AND NUCLEAR MEDICINE (ISBN I 870965 26 4) 416

pp, Hardback, Proceedings of a Workshop, Wurzburg, June 1992, NEUTRON DOSIMETRY 498 pp. Softback, Proceedings of the Sixth Price 80.00 Symposium on Neutron Dosimetry, Neuherberg, October 1987, Price 75.00

MICRODOSIMETRY (ISBN 1 870965 21 3) 500 pp, Hardback, NATURAL RADIOACTIVITY 560 pp, Softback, Proceedings of the Proceedings of the Eleventh Symposium on Microdosimetry, Gadmburg, Fourth International Symposium on the Natural Radiation Environment, September 1992, Price 90.00 Lisbon, December 1987, Price 85.00

DECISION MAKING SUPPORT FOR OFF-SITE EMERGENCY BIOLOGICAL ASSESSMENT OF OCCUPATIONAL EXPOSURE TO MANAGEMENT (ISBN I 870965 25 6) 320 pp, Hardback, Proceedings ACTINTOES 400 pp. Softback, Proceedings of a Workshop, Versailles, of a Workshop, Schloss Elmau, October 1992, Price 60.00 May 1988, Price 65.00

INTAKES OF RADIONUCLIDES - DETECTION, ASSESSMENT AND IMPLEMENTATION OF DOSE-EQUTVALENT QUANTITTES ESTO LIMITATION OF OCCUPATIONAL EXPOSURE (ISBN 1 870965 28 0) RADIATION PROTECTION PRACTICE (ISBN 1 870965 03 5) 166 pp, 370 pp, Hardback, Proceedings of a Workshop, Bath, September 1993, Softback, Proceedings of a Seminar, Braunschweig, June 1988, Price 80.00

Price 35.00 INDIVIDUAL MONITORING OF IONISING RADIATION - THE

IMPLEMENTATION OF DOSE-EQUIVALENT METERS BASED ON IMPACT OF RECENT ICRP AND ICRU PUBLICATIONS. MICRODOSIMETRIC TECHNIQUES (ISBN 1 870965 04 1) 156 pp, (ISBN 1 870965 29 9) 232 pp, Hardback, Proceedings of a Workshop, Softback, Proceedings of a Seminar, Schloss Elmau, October 1988, Villigen, May 1993, Price 45.00

Price 30.00 INDOOR RADON REMEDIAL ACTIONS -THE SCIENTIFIC AND

MICRODOSIMETRY (ISBN 1 870965 05 X) 460 pp, Softback, PRACTICAL IMPLICATIONS. (ISBN 1 870965 30 2) 400 pp, Hardback, Proceedings of the Tenth Symposium on Microdosimetry, Rome, May 1989, Proceedings of a Workshop, Rimini, Italy, June 27 to July 2 1993,

Price 80.00 Price 90.00

STATISTICS OF HUMAN EXPOSURE TO IONISING RADIATION (ISBN 1 870965 08 6) 280 pp, Hardback, Proceedings of a Workshop, FUTURE PUBLICATIONS Oxford, April 1990, Price 60.00

ADVANCES IN RADIATION MEASUREMENTS - APPLICATIONS RESPIRATORY TRACT DOSIMETRY (ISBN 1 870965 09 4) 268 pp, AND RESEARCH NEEDS IN HEALTH PHYSICS AND DOSIMETRY Hardback, Proceedings of a Workshop, Albuquerque, July 1990, (ISBN 1 870965 33 7) 450 approx pp, Hardback, Proceedings of a

Price 60.00 Workshop, Chalk River, October 1994, Price 90.00

Please send your orders to: Nuclear Technology Publishing, P.O. Box No 7, Ashford, Kent TN23 1YW, England

Telephone 44 (0) 1233 641683 Fax 44 (0) 1233 610021 EEC-PUB.J 05.12.94

Radiation Protection Dosimetry Vol. 57, Nos. 1 - 4 , pp. vii viii (1995) Nuclear Technology Publishing

Editorial Data Analysis in Quality Control and Radiation Protection in Diagnostic Radiology and Nuclear Medicine

The Workshop on 'Data Analysis in Quality Control and Radiation Protection of the Patient in Diagnostic Radiology and Nuclear Medicine' was conceived as a forum for the evaluation of ten years of effort in these fields.

For the co-organisers, the Department of Medical Physics of the Hospital Santa Maria della Misericordia, Udine, the World Health Organization and the Commission of the European Communities, this evaluation was seen from different points of view. The research group of Udine, as probably did most of the participants, wanted to check on progress since the seminar on 'Criteria and Methods for Quality Assurance in Medical X ray Diagnosis' held ten years ago in Udine and on which their current studies in the evolution of the field were based. For the World Health Organization the Workshop should help to identify means for the achievement of the overall goal good quality of medical services for everybody. The Commission of the European Communities hoped to assess the effectiveness of quality assurance measures in implementing the requirements of the Council Directive on the radiation protection of the patient (84/466 EURATOM, 4 September 1984).

About 100 papers involving some 300 research workers addressed detailed objectives of the Workshop including:

- overviews of trends and evolution in the past decade, - intercomparisons of dose reduction concepts, methods and practices, - definition of procedures for optimisation of radiation protection, as regards the clinical, equipment and organ

isational aspects as well as training, - analysis of the relationship between cost, risk and benefit, - perspectives of transfer of know-how to the user. For a greater coherence of the subject areas dealt with, the various contributions have been grouped slightly

differently in these Proceedings from that occurring during the Workshop, thus emphasising the following main topics of interest, namely Guidance, European and Multi-national Trends and Activities, Image Quality, Computer Applications, (subdivided into knowledge based systems, and models and tools), Quality Control, (subdivided into instruments and methods, and surveys), Radiation Protection of the Patient, (subdivided into dose evaluation methods, dose reduction and surveys).

For a research area that is in full evolution it was evident that for every question answered, two new ones emerged. The chairpersons of the Workshop sessions had therefore been invited to guide the session discussions in a way that answers to the following questions could be formulated:

- What radiation protection strategies should be defined for diagnostic radiology and nuclear medicine? - What optimisation strategies can be recommended? - What is the practical relevance of the assessment of doses to the patient? - What socio-economic factors (health care concepts and spending, industrial activities etc) should be incorp

orated in radiation protection strategies? - What are the education and training priorities in radiation protection in diagnostic radiology and nuclear med

icine? - What are the most important specific research priorities in all these items? The discussions taking place in the various Workshop sessions have been summarised by the chairpersons with

these questions in mind and are reproduced in the penultimate section, followed by the final discussion and conclusions of the Workshop.

The final discussion emphasised the need for better understanding of some specific components of radiation protection during radiological imaging procedures. In particular, since the reduction of patient dose is constrained by the need to obtain images of sufficient diagnostic content, it is necessary to understand the relationship between diagnostic content, physical measures of image quality and patient dose. There was a consensus that this task requires a multidisciplinary approach in order to lay down explicit links between the medical and the radiation protection requirements. Quality assurance programmes, which include periodic patient dose measurements and comparison with reference doses, can provide an essential feedback mechanism on the level of protection achieved. The CEC initiative to define, develop and test practical image quality and patient dose criteria at a European level

vii

EDITORIAL for common examinations01, was considered to be a valuable effort in this direction.

It was encouraging to find that this conceptual framework for optimisation, including quality criteria and technical performance requirements, is being dealt with from different points of view at the local, regional, and national level, as well as by the various competent professional bodies and international organisations such as ICRP, ICRU, WHO and UNSCEAR.

It was the wish of the organisers of the Workshop that the evaluation of the data collected during the last decade on quality control and radiation protection become a milestone in the field of quality assurance in diagnostic radiology and nuclear medicine, against which further progress can be evaluated. As with the preceding six European Workshops dealing with quality assurance in medical radiology, this one has also contributed towards a common terminology and provided fertile ground in which future research can be rooted. It is hoped that these Proceedings will consolidate this purpose.

REFERENCE . CEC. Quality Criteria for Diagnostic Radiographic mages. Working Document CEC 7173/90, 2nd Edition,

Commission of the European Communities, Brussels, June 1990. G. Contento

. Wall . Schibilla

D. Teunen

vin

Radiation Protection Dosimetry Vol. 57, Nos. 1 -4 , pp. ix-xv (1995) Nuclear Technology Publishing

Contents Contents

Opening Address

J. Sinnaeve 1

GUIDANCE

Referral Criteria for Selection of Patients and Diagnostic Procedures L. Dalla Palma, C. Ricci and S. Magnaldi (INVITED PAPER) 3 The Scientific Work-Up of Radiographie Image Quality Now and a Decade Ago - The Radiologist's Approach J.W. Oestmann (INVITED PAPER) 9

Trends in X Ray Diagnosis and Nuclear Medicine F.E. Stieve (INVITED PAPER) 13

Interlaboratory Comparison of Imaging Devices in Nuclear Medicine H. Bergmann (INVITED PAPER) 21

WHO and Rational Reduction of Patient Dose G. Hanson (INVITED PAPER) 27

EUROPEAN AND MULTINATIONAL TRENDS AND ACTIVITIES

Initiatives, Achievements and Perspectives with regard to the Council Directive of September 3 1984 laying down Basic Measures for Radiation Protection of Persons undergoing Medical Examination or Treatment D. Teunen and National Experts: A. Wambersie (Belgium), O. Hjardemaal (Denmark), A. Costa (France), B. Bauer (Germany), D. Panagiotis (Greece), G. O'Reilly (Ireland), F. Mazzei andM. Paganini Fioratti (Italy), C. Back (Luxembourg), T. Zoetelief (the Netherlands), A.F. Carvalho (Portugal), E. Vano (Spain), and S. Ebdon-Jackson (United Kingdom) (INVITED PAPER) 33

Initiatives of EFOMP in the Field of Radiation Protection and Quality Assurance A. Del Guerra (INVITED PAPER) 73

Radiation Protection Education and Training of Radiographers P. Elsakkers (INVITED PAPER) 77

Practical Impact of the Evolution and Changes of ICRP Recommendations on Radiological Protection in Medicine S. Mattsson and A. Almn (INVITED PAPER) 79

UNSCEAR Data Collections on Medical Radiation Exposures: Trends and Consequences J. Valentin (INVITED PAPER) 85

IEC Projects and Achievements in the Field of Quality Assurance in Diagnostic Radiology W. Puschert (INVITED PAPER) 91

IX

CONTENTS Results of the IAEA-CEC Co-ordinated Research Programme on Radiation Doses in Diagnostic Radiology and Methods for Reduction P. Ortiz, C. Maccia, R. Padovani, E. Vano, G. Alm Carlsson and H. Schibilla 95

The United States Experience in Patient Dose and Image Quality .H. Suleiman, B.J. Conway, F.G. Rueter, J.L. McCrohan, R.J. Slayton andR.G. Antonsen (INVITED PAPER) 101

IMAGE QUALITY

CEC Quality Criteria for Diagnostic Radiographic Images - Basic Concepts B.M. Moores (INVITED PAPER) 105

The 1991 CEC Trial on Quality Criteria for Diagnostic Radiographic Images C. Maccia, M. Ariche-Cohen, X. Nadeau and C. Severo (INVITED PAPER) Ill

Evolution of Quality Assurance in Paediatric Radiology K. Schneider (INVITED PAPER) 119

Quality Control and Image Quality Criteria in Computed Tomography J. Albrechtsen, J. Hansen, L.C. Jensen, K.A. Jessen and A.G. Jurik 125

Patient Dose and Image Quality for Computed Tomography in Several Dutch Hospitals J. Geleijns, J.J. Broerse, J. Zoetelief, D. Zweers and J.G. van Unnik 129

Image Quality and Dose Distribution in Multiscan and Fast Ring CT Systems F. Levrero, G.L. Coscia, A. Pilot, F. Cavagnetto and E. Zucchi 135

Image Quality Criteria Applied to Digital Radiography HP. Busch, K. Faulkner and J.F. Malone 139

Standard Tools for Analysis of Image Quality in Digital Imaging N. Meier and M. Fiebich 141

A Novel View on Detail Perception and on the Influence of the Film System Parameters R. Bollen 145

Optimisation of Image Quality in Mammography K.J. Robson, C.J. Kotre and K. Faulkner 151

Impact of Several Recommended Actions for Improving the Image Quality in Mammography M. Chevalier, P. Moran and E. Vano 155

Quality Assurance and Patient Dosimetry in Mammography: a Retrospective I.A. Castellano, D.R. Dance, R. Davis, S.H Evans, C.H. Jones and C.A. Parsons 159

Survey of Mammographie Image Quality in the UK K.C. Young, M.L. Ramsdale and A.G. Rust 163

Image Quality and Patient Dose in Diagnostic Radiology D. Saure, G. Hagemann and H. Stender 167

CONTENTS Assessment of Image Quality for Chest Radiography in the West Midlands H.M. Warren-Forward and J.S. Millar 171

COMPUTER APPLICATIONS

Knowledge Based Systems

Data Analysis and Information Modelling: Objects, Codes, Concepts D.P. Pretschner (INVITED PAPER) 175 Knowledge Based Approach to Quality Control in Diagnostic Radiology G. Contento, R. Padovani, V. Roberto, O. Varin, and C. Della Giusta (INVITED PAPER) 185

Knowledge Acquisition, Representation and Reasoning in a Gamma Camera Quality Control Expert System P.J. Slomka, T.D. Cradduck and L.A. Chudziak 191

Models and Tools

Medical Devices Quality Assurance Services A. Talbot, M. Kragsholm, A. Mnsson and S.S. Nielsen 195

Quality Control in Mammography Based on Automatic Acquisition of Exposure Parameters T.P. Walderhaug, G. Einarsson, S. Kristinsson, S.M. Magnusson and B.F. Sigsson 199

Computer Applications in Radiation Protection P.R. Cole and B.M. Moores 203

Optimisation of the Design of Antiscatter Grids by Computer Modelling D.R. Dance, M. Sandborg, G. Alm Carlsson and J. Persliden 207

Results from an Optimisation of Grid Design in Diagnostic Radiology M. Sandborg, D.R. Dance, G. Alm Carlsson and J. Persliden 211

Computer Model for RiskBenefit Analysis of Mammographie Breast Cancer Screening J.Th.M. Jansen and J. Zoetelief 217

QUALITY CONTROL

Instruments and Methods

In Field Testing of a New Instrument for Quality Control in Mammography M. Gambaccini, M. Marziani, CG. Candini, E. De Guglielmo and C. Bertaggia 221 A Noninvasive Method to Control the Tube Current Calibration of Diagnostic Radiology Equipment N.J. Uys, C.P. Herbst, A. van Aswegen, M.G. Latter, M.A. Sweetlove and J.F.K. de Villiers 227

A Compound Hard Tissue Phantom in Radiologic Diagnostic Optimisation O. Eckerdal, E. Helmrot and G. Aim Carlsson 231

Optimisation of Radiographic AEC System Performance J. Upton, B. Moran and J.F. Malone 237

CONTENTS Quality Assurance Programme Applied to Mobile X Ray Equipment B. Tuohy, G. Tuohy, P. Cooney, B. Moran and J.F. Malone 241

Contrast-Detail Testing Techniques for Modern X Ray Image Intensifier Systems CJ. Kotre, N.W. Marshall and K Faulkner 245

A Review of the Background to the Decision to Write-off Fluoroscopy Equipment in 15 Instances, - and the Impact of Patient Dose and Image Quality in Practice

J.F. Malone, P. Cooney, HP. Busch and K Faulkner 249

A Simple Test for X Ray Tube Filtration R.E. Gallini, S. Belletti, V. Berna, U. Giugni and G. Prandelli 253

Determination of the Speed of Medical Radiography Screen-film Systems F.R. Verdun, F. Bochud, J.F. Valley and O. Lm Thanh 257

A Practical Approach to a Quality Assurance Programme for Radiography at the Constancy Check Level B. Moran, J. Upton, P. Cooney and J.F. Malone 263

Automatic Exposure Control in Fluoroscopic Imaging P. Cooney, D. M. Marshand J.F. Malone 269

Measurement of Wiener Spectra in Digital Systems D.M. Marsh, P. Cooney, B. P. McMahon and J.F. Malone 273

An Assessment of the Variations in Image Quality with Multiformat Cameras D.M. Marsh, J.F. Malone and P. Cooney 277

Pitfalls in the Use of Flood Sources "Co E. Busemann Sokole, A. Kugi and H Bergmann 281

NIR: A Database of All X Ray Units in Use in Lower Saxony to Improve Radiation Protection and Quality Control

H. Brggemeyer and T. Siewert 285

Surveys

Some Results from a Diagnostic Radiology Optimisation Programme in the Madrid Area E. Vano, L. Gonzales, E. Guibelalde, J.M.Fernndez, A. Calzado and M.J. Ruiz 289 Quality Control in Conventional Diagnostic Radiology in Greece V. Neofotistou, M. Molfetas and N. Panagiotakis 293

Quality Control and Patient Dose for X Ray Examinations in Some Hospitals in Estonia A. Servomaa, S. Rannikko, T. Parviainen, P. Holmberg, E. Kuus, T. Mrsepp and V. J rv 297

What is the True Film-Screen Sensitivity H.M. Warren-Forward 301

Survey of Image Viewing Conditions for X Ray Film P. Cooney, C Davis, J.B.Y. Chua, W. vander Putten and J.F. Malone 305

Xli

CONTENTS A Quality Assurance Programme for Medical X Ray Diagnostic Units Carried Out in Belgium D. Godechal, J. Delhove, C Mambour, J. CoomansandA. Wambersie 309

Quality Control in the Radiological Departments of the Florence General Hospital C. Gori, G. Belli, S. Calvagno, L. Capaccioli, A. Guasti, G. Span and G. Zate Ui 315

The RTI DIGI-X Plus QA System in Routine Practice M. Princivalli, L. Stea, P.L. Ordonez, L. Bussoli and C Marchetti 317

Quality Control in Mammography: the Pilot Campaign of Breast Screening in the Bas-Rhin Region C Maccia, X. Nadeau, R. Renaud, S. Castellano, P. Schaffer, R. Wahl, P. Haehnel, G. Dale and B. Gairard 323

Analysis of the Results of a QC Project on Mammography in Greece A. Flioni-Vyza, S. Xenofos, G. Panagiotakis, E. Giakoumakis and B. Proimos 329

Quality Control in Computed Tomography Performed in Portugal and Denmark A.F. Carvalho, A.D. Oliveira, J. Alves, J. V. Carreiro, L.C. Jensen and KA. Jessen 333

Radiation Protection Problems with Dental Radiological Equipment P. Cooney, G. Gavin, J. Rajan and J.F. Malone 339

Daylight Film Processor/Loader Systems - Aspects of Performance and Distribution J. Upton, B. Moran, M. Duggan and J.F. Malone 343

Implementation of a Quality Control Programme at the Departments of Radiology and Nuclear Medicine of the Netherlands Cancer Institute (NKI-AvL) S.H Muller 347

Quality Control of Nuclear Medicine Instruments in Argentina

M. Levi de Cabrejas and C. Giannone 351

RADIATION PROTECTION OF THE PATIENT

Dose Evaluation Methods

National Patient Dose Measurement Protocols: an Investigation on Behalf of ICRU P.J. Roberts (INVITED PAPER) 355 Patient Dose Protocols and Trends in the UK B.F. Wall and P.C. Shrimpton 359

Dose-Area Product and Body Doses N. Petoussi-Hen, W. Panzer, M. Zankl and G. Drexler 363

A Comparison of Two Methods for Estimating Effective Dose in Abdominal Radiology N.W. Marshall, K Faulkner, HP. Busch, D.M. Marsh and H Pfenning 367

Effective Doses for Different Techniques Used for PA Chest Radiography F.W. Schultz, J. Geleijns and J. Zoetelief 371

Computed Tomography Dose Assessment - a Practical Approach W. Leitz, B. Axelsson and G. Szendr 377

xiii

CONTENTS A Comparison of Measured and Calculated Organ Doses from CT Examinations A. Calzado, S.R. Sanz, M. Melchor and E. Vano 381

Dose Profile and Dose Index Analysis in Computed Tomography A.D. Oliveira, J. G. Alves, A.F. Carvalho and J. V. Carreiro 387

Organ Doses for Children from Computed Tomographic Examinations M. Zankl, W. Panzer, N. Petoussi-Hen and G. Drexler 393

Calculation of Air Kerma to Average Glandular Tissue Dose Conversion Factors for Mammography J. Zoetelief and J. Th.M. Jansen 397

Comparison of Dose Measurement Protocols in Mammography K.C Young 401

An Investigation into Variations in the Estimation of Mean Glandular Dose in Mammography K. Faulkner and K Cranley 405

Physiologically Based Pharmacokinetics Model for Estimating Urinary Excretion of Short Half-life Nuclides in Nuclear Medicine K. Akahane, M. Kai, E. Konishi, T. Kusarna and Y. Aoki 409

Dose Reduction

The Increasing Importance of X Ray Computed Tomography as Source of Medical Exposure P.C. Shrimpton and B.F. Wall 413

Intensive Care Department: Evaluation of the Radiological Activity and Criteria for Reduction of Patient and Worker Exposure S. Magri, S. Arisi, S. Camerini, M. Nolli, P.U. Marini and G. Rozzi 417

An Initial Report on the Investigation of High Patient Doses for the Lateral Lumbosacral Projection in the Lumbar Spine Examination B. Moran, J. Upton, M. Rafferty, G. Boyle and J.F. Malone 423

Recommended Actions for Reduction of Breast Doses in the Area of Madrid. An Evaluation. P. Moran, M. Chevalier and E. Vano 429

Evaluation of Radiation Exposure to Personnel in Cardiac Angiography B. Axelsson, T. Cederlund and B. Svane 433

Surveys

Comparison of Entrance Surface Doses and Radiographic Techniques in the West Midlands (UK) with the CEC Criteria, Specifically for Lateral Lumbar Spine Radiographs E.A. McNeil, D.E. Peach and D.H. Temperton 437

Patient Dosimetry During Chest Radiography H.M. Warren-Forward 441

Patient Dose Measurement and Dose Reduction in East Anglia (UK) J.P. Wade, KE. Goldstone and P.P. Dendy 445

XIV

CONTENTS Patient Doses and Radiation Risks in Filmscreen Mammography in Finland A. Servomaa, T. Parviainen and T. Komppa 449

Level and Distribution of the Radiation Dose to the Population from a Mammography Screening Programme in New Zealand S.M. Bulling and J.J. Nicoli 455

Intestinal Biopsy in Children with Coeliac Disease; a Swedish National Study of Radiation Dose and Risk J. Persliden, H.B.L. Pettersson and K. FlthMagnusson 459

Dose Distribution in Children at Chest Radiography A. Almn and S. Mattsson 463

Computed Tomography Practice in Sweden. Quality Control, Techniques and Patient Dose G. Szendr, B. Axelsson and W. Leitz 469

Data Analysis on Patient Exposures in Cardiac Angiography J.C Huy skens and A.W. Hummel 475

Radiation Risk and Exposure of Radiologists and Patients during Coronary Angiography and Percutaneous Transluminal Coronary Angioplasty (PTCA) J. Karppinen, T. Parviainen, A. Servomaa and T. Komppa 481

Radiation Exposure of Patients by Using Modern Digital Fluoroscopy Systems J. Kopp, W. Maier and C Losereit 487

Absorbed Dose to the Skin in Radiological Examinations of Upper and Lower Gastrointestinal Tract G. Zonca, A. Brusa, M. Bellomi, G. Cozzi, A. Severini, . Somigliano, C. Pasqualotto and A. Sichirollo . 489

Discussions by Session Chairpersons 493

Final Discussion and Conclusions 503

Author Index 507

List of Participants 508

Radiation Protection Dosimetry is abstracted or indexed in APPLIED HEALTH PHYSICS ABSTRACTS AND NOTES, Chemical Abstracts, CURRENT CONTENTS, Energy Information Abstracts (Cambridge), EXCERPTA MEDICA (EMBASE), Health and Safety Science Abstracts (Cambridge), INIS AOMINDEX (hard copy and CD ROM), INSPEC, Nuclear Energy (Czech Republic), QUEST and Referativja Zhurnal.

xv

Nuclear Technology Publishing 21 Years 1974 - 1995

1995 Journal Prices APPLIED HEALTH PHYSICS ABSTRACTS AND NOTES (1995 Volume 21)

ISSN 0305-7615 (commenced publication 1974)

An international abstracts journal in applied health physics covering radiation protection, radiation dosimetry, measurement techniques, radiation effects and radiation accidents. The journal is published quarterly and includes approximately 2,500 abstracts per year.

1995 subscription cost: 160.00 (UK), US$330.00 (outside UK)

********************

RADIATION PROTECTION DOSIMETRY (1995 Volumes 57, 58, 59, 60, 61 and 62) ISSN 0144-8420 (commenced publication 1981)

An international journal covering all aspects of personal and environmental dosimetry and monitoring for ionising and non-ionising radiations, including biological aspects, physical concepts, external dosimetry and monitoring, internal dosimetry and monitoring, environmental and workplace monitoring and dosimetry related to patient protection. There are six volumes of four issues per volume for 1995.

1995 subscription cost: 550.00 (UK), US$1190 (outside UK)

*******************

INTERNATIONAL JOURNAL OF RADIOACTIVE MATERIALS TRANSPORT (1995 Volume 6) ISSN 0957-476X (commenced publication 1990)

An international journal covering all aspects of transport of radioactive materials including regulations, package design, safety assessments, package testing, transport experiences and accidents. The journal is published quarterly and includes regular news features from the International Atomic Energy Agency.

1995 subscription cost 96.00 (UK), US$195 (outside UK)

******************** 1995 PUBLICATIONS

QUALITY CONTROL AND RADIATION PROTECTION OF THE RADIOBIOLOGY OF INHALED RADIONUCLIDES PATIENT IN DIAGNOSTIC RADIOLOGY AND NUCLEAR MEDICINE (ISBN 1 870965 38 8) 100 approx. pp. Softback. Proceedings of a Symposium. (ISBN 1 870965 37 X) 450 approx. pp. Hardback. Proceedings of a Workshop, Richland. Washington, November 1993. Price 25.00 Grado, October 1993, Price 90.00

BRITTLE FRACTURE SAFETY ASSESSMENT ADVANCES IN RADIATION MEASUREMENTS - APPLICATIONS AND (ISBN I 870965 40 X) 100 approx. pp. Softback, Proceedings of a Workshop, RESEARCH NEEDS IN HEALTH PHYSICS AND DOSIMETRY Krefeld, October 1994. Price 24.00 (ISBN 1 870965 33 7) 450 approx. pp. Hardback, Proceedings of a Workshop. Chalk River. October 1994, Price90.O0 RADIOBIOLOGICAL CONSEQUENCES OF NUCLEAR ACCIDENTS and

NUCLEAR ACCIDENTS, RADIOECOLOGY AND HEALTH APPLICATIONS OF COMPUTERISED GLOW CURVE ANALYSIS TO (ISBN 1 870965 41 8) 250 pp approx. Hardback. Proceedings of joint Workshops THERMOLUMINESCENCE DOSIMETRY held in Moscow, October 25 - 28 1994. Price to be announced. (ISBN 1 870965 36 I) 100 approx. pp. Softback. Price 25.00

These items are published as part of.thc journal RADIATION PROTECTION DOSIMETRY, and provided to subscribers as part of the annual subscription. These items are published as part of THE INTERNATIONAL JOURNAL OF RADIOACTIVE MATERIALS TRANSPORT and are provided to subscribers as part of the annual subscription.

a**** * * * A free cumulative index for the first 50 volumes of RADIATION PROTECTION DOSIMETRY (1981 -1993) is available now. Please send for your copy, quoting the reference number at the foot of this advertisement. A complete catalogue of all of our radiation protection publications will be available early in 1995.

a * * * * * *

Please send your orders to: Nuclear Technology Publishing P.O. Box No 7, Ashford

Kent TN23 1YW, England PRICES-12. Telephone 44 (0) 1233 641683 Fax 44 (0) 1233 610021

WSA

l' M B

H Q

Radiation Protection Dosimetry Vol. 57, Nos. 1-4, pp. 1-2 (1995) Nuclear Technology Publishing

OPENING ADDRESS J. Sinnaeve Commission of the European Communities Radiation Protection Research Action

The Commission of the European Communities organised its first Workshop on Quality Assurance in Diagnostic Radiology not far from here in Udine, in 1984, the famous year of Orwell's visions of the future. One way of evaluating the achievements in Quality Assurance in Diagnostic Radiology is to assess whether the objectives of 10 years ago have been realised. We were therefore very pleased that we could convince the Udine team of the Servizio di Fisica Sanitaria to co-organise a Workshop with us once more. I am particularly grateful to them and especially to Dr Gilberto Contento who has shouldered the enormous amount of work involved in the organisation of the Workshop.

I thus have great pleasure in welcoming you to the Workshop which will analyse and evaluate the data that have been produced in the past decade on Quality Control and Radiation Protection of the Patient in Diagnostic Radiology and Nuclear Medicine. The objective is to define together what are the most appropriate methods for both analysis and evaluation, and to check how far the achievements have fulfilled our objectives.

The Commission's services are periodically requested to evaluate the scientific and practical value of the research programmes of the CEC and our experience is that this stimulates creative criticism and opens up new ways of thinking. The Commission's Radiation Protection Actions have been and continue to be evaluated by external experts, who have recognised the need to deploy special efforts on radiation protection of the patient; but up to now we have not called upon the various professionals involved that is to say radiologists, radiographers, medical physicists, radiation protection surveyors and teachers, health authorities and manufacturers, as well as international organisations to evaluate together our efforts, our achievements and our objectives in this important field of radiation protection, related to the medical use of ionising radiation. This part of radiation protection research and legislation concerns every member of the public as a potential patient. The coherence of our work and its results should therefore be ensured all over Europe and contribute to worldwide coherence. We have also come together here in order to look for ways and means to improve this coherence.

The seven Workshops and Seminars organised by the Commission demonstrate clearly where our main interests lay over the past decade.

We started with the possibilities for dose reduction

in diagnostic radiology; we defined concepts for quality assurance programmes; we evaluated the technical and physical parameters for quality control; we established quality criteria for diagnostic radiographic images and their link to radiographic techniques and doses to the patient, in particular with respect to certain conventional and digital radiographic procedures; we discussed the role of dosimetry as well as test objects and phantoms for quality control measures' '~7).

The conclusive discussions of all of these events highlighted the achievements and requirements formulated for future actions and research efforts. It is an instructive read I can recommend it to all of you if we wish to find out whether our aims were realistic, whether we concentrated on the most relevant items and whether we succeeded in proposing the most appropriate radiation protection means and measures for the use of ionising radiation in medicine.

There is, however, some discrepancy between the radiation protection measures that have been recommended and the public concern over radiation hazards linked to the medical use of ionising radiation and radionuclides. Several newspapers and magazines have recently denounced the possible dangers involved after accidents had occurred due to human errors and equipment failure.

In this context our evaluation should also examine: (i) Whether the knowledge available of radiation pro

tection measures is being properly applied? (ii) Whether this knowledge is suitable for medical-

clinical priorities and circumstances? (iii) What else we need to know in order to control the

true situation? We must establish and seek agreement on the criteria that will enable us to study these questions.

At the end of the Book of Abstracts a series of Open Questions is listed for the Final Discussion. The answers that we might find together here should guide us in the effort to establish those criteria and also to establish the strategies needed to facilitate their application. I would like to stress how serious these questions are also for the definition of our future goals for research and legislation, including the revision of the Council Directive laying down basic measures for the radiation protection of persons undergoing medical examination or treatment.

Research and Training on radiation protection for the medical use of ionising radiation and radionuclides must

J. SINNAEVE certainly concentrate more on the efficiency of radiation protection measures established so far, on their compatibility with medical requirements and on their links to overall optimisation of diagnosis and health care. For the definition of future priorities for radiation protection we must therefore study the ways and means by which the criteria for Quality Assurance and Quality Control can be improved; with special emphasis on the definition of image quality in a most objective way: this might be achieved in promoting new lines of research, such as imaging sciences, the creation of new branches of radiological sciences and the development of new radiological methods and diagnostic tools.

We will not, of course, be able to find answers to all these questions, nor will we have the necessary financial means to follow up all suggestions. However, at a moment where everybody is being invited worldwide to think more rationally about the use of resources and manpower, we must coordinate our efforts in order to define rational strategies on how to fulfil the most urgent tasks, even if this can only be done in small steps.

I am impressed by the number of presentations that will take place: about 120 oral and poster communications, involving more than 300 research workers, will give a picture of the research results available in some 20 European and nonEuropean countries. They will contribute to the definition and analysis of radiation protection problems. They will also provide the material for the training programmes of the coming generation and that is one of the most rewarding aspects of all our

efforts, in my opinion. We must recognise our responsibility in transmitting our commitment in this field of research to young scientists. The CEC understands that one of its main tasks in research is to prepare the ground for sound research work to be carried out and guarantee the continuity of European expertise in selected priority areas.

I very much welcome the organisation of the Round Table of the national experts who will review the achievements in radiation protection of the patient in their countries, indicate which research areas are being concentrated on and show how the implementation of radiation protection is facilitated on a more individual basis, so that we can learn from each other with the aim of approaching the optimum level. In addition, a number of international bodies will summarise their concepts and goals concerning the various aspects of radiation protection in medicine and will thus add new criteria for the evaluation of the outcome of the current research work.

We have progressed in this field to a point where we are now closer to the possible standardisation of quality assurance and quality control measures in practice. Standardised measures will enable us to evaluate the efficiency of radiation protection of the patient.

With this in mind I would like to invite you to benefit from this three day meeting by creating new links across the various professions, subject areas and geographical distances.

REFERENCES 1. Patient Exposure to Radiation in Medical Xrays Possibilities for Dose Reduction. Proceedings of a Seminar jointly organ

ised by the CEC and the Gesellschaft fr Strahlen und Umweltforschung, MunichNeuherberg (FRG), 2730 April 1981. Eds G. Drexler, H. Eriskat and H. Schibilla. Report EUR 7438 EN (1981).

2. Criteria and Methods for Quality Assurance in Medical Xray Diagnosis. Proceedings of a Scientific Seminar organised jointly by the CEC and the Centro di Ricerca Applicatae e Documentazione, held in Udine, (I), 1719 April 1984. Eds G. Drexier, H. Eriskat and H. Schibilla. Report EUR 9255 EN, BIR Suppl. 18 (1985).

3. Technical and Physical Parameters for Quality Assurance in Medical Diagnostic Radiology: Tolerances, Limiting Values and Appropriate Measuring Methods. Proceedings of a Workshop organised by the CEC, held in Brussels (B), 2325 February 1988. Eds B.M. Moores, F. E. Stieve, H. Eriskat and H. Schibilla. Report EUR 11620 EN, BIR Report 18 (1989).

4. BIR. Optimization of Image Quality and Patient Exposure in Diagnostic Radiology. Proceedings of a Workshop organised jointly by the CEC and the National Radiological Protection Board Chilton, held in Oxford, (UK), 2729 September 1988. Eds B. M. Moores, . F. Wall, . Eriskat, . Schibilla. Report EUR 11842 EN, BIR Report 20 (1989).

5. Dosimetry in Diagnostic Radiology. Proceedings of a Seminar, jointly organised by the CEC, the PhysikalischTechnische Bundesanstalt, Braunschweig (FRG), the World Health Organization and the International Commission on Radiation Units and Measurements held in Luxembourg (L), 1921 March 1991. Eds H. M. Kramer and K. Schmier. Report EUR 14180 EN. Radit. Prot. Dosim. 43(14) (1992).

6. Digital Radiography: Quality Assurance and Radiation Protection. Proceedings of a Workshop jointly organised by the CEC, the Klinikum Mannheim and the European Association of Radiology, held in Mannheim (FRG), 79 May 1992. Eds H. P. Busch and M. Georgi (SchnetztorVerlag GmbH, Konstanz) EN (1992).

7. Test Phantoms and Optimisation in Diagnostic Radiology and Nuclear Medicine. Proceedings of a Workshop jointly organised by the CEC, the Forschungszentrum fr Umwelt and Gesundheit, Neuherberg (FRG), the International Commission on Radiation Units and Measurements and the European Federation of Organisations for Medical Physics, held in Wrzburg (FRG), 1517 June 1992. Eds B. M. Moores, . Petoussi, . Schibilla and D. Teunen. Report EUR 14767 EN. Radial. Prot. Dosim. 49(13) (1993).

Radiation Protection Dosimetry Vol. 57, Nos. 14, pp. 38 (1995) Nuclear Technology Publishing

REFERRAL CRITERIA FOR SELECTION OF PATIENTS AND DIAGNOSTIC PROCEDURES L. Dalla Palma, C. Ricci and S. Magnaldi Universit degli Studi di Trieste, Istituto di Radiologia Ospedale di Cattinara, 134149 Trieste, Italy

INVITED PAPER Abstract In order to maximise the benefits of modem diagnostic procedures and to minimise costs both for the health service and the individual, criteria for selecting patients undergoing diagnostic procedures must be carefully established. This review analyses the main factors (related both to patients and technology) which should influence our diagnostic choices. The general meaning of efficacy, efficiency, cost and benefits in a radiological milieu is defined and then applied to a number of highly practical situations where new technologies (ultrasonography, computed tomography and magnetic resonance) have revolutionised diagnostic approaches in the past decade. The review emphasises that referral criteria must be founded on a 'first priority' basis so that decisions as to diagnostic protocol are taken on the basis of clinical features, epidemiological data and simple laboratory tests; the consequent steps in diagnostic protocols are influenced by a priori knowledge of efficacy, efficiency, prognostic and therapeutic value of the diagnosis and cost considerations.

INTRODUCTION The increasing number of radiological examinations

in the population as a whole, the increasing cost for health care and the advent and most appropriate use of new technologies which do not use ionising radiations are important issues in radiology today.

Prior to any discussion of this topic it is useful to single out some keywords which help in understanding how to draw up guidelines for optimal use of imaging modalities.

Accuracy: often used as a synonym of efficacy this term represents the ratio between actual and ideal diagnostic performance'". In diagnostic imaging it can be defined as the ability to diagnose correctly true positive (TP) and true negative (TN) cases and is given by the following ratio: TP + TN 100/grand total. It is, however, also useful in modern diagnostic protocols in order to characterise the role of a given modality, to define other parameters such as sensitivity (TP/TP+FN) and specificity (TN/TN + FP)(2>.

Efficiency may be defined as the ratio between the resources which should ideally be employed and those which are actually used"1. Referred to diagnostic imaging it represents the ratio between accuracy and cost. The modality which offers the highest accuracy at the lowest cost is highly efficient.

Cost is an increasingly important factor which is sometimes difficult to quantify unless we consider purely financial cost, usually expressed as cost/exam/year and is related to initial investment for the technology and to break even point. In reality, there is an additional optional cost regarding initial and inservice training of personnel.

Biological cost is considered separately because it is

one of the most important issues in modern diagnostic imaging, involving both radiologists and patients. This factor is mainly related to exposure to ionising radiations of many categories of subjects but particularly children, pregnant women and adults of reproductive age. Whereas this cost is ideally zero for modalities such as ultra sonography (US) and magnetic resonance (MR), it can be only estimated for modalities like conventional radiology, CT and digital subtraction angiography (DSA).

In recent years US has enabled this cost to be reduced in diagnostic imaging, by replacing X rays in several situations (for guiding diagnostic and therapeutical manoeuvres as in biopsies of abdominal organs, drainage of abscesses, percutaneous nephrostomies and gastrostomies, and even portosistemic shunts). Recently, US has also become an important tool for studying dynamic phenomena in areas where the reduction of dose is crucial, such as the female pelvis.

There is a further biological cost in the risk associated with the use of contrast media (quantifiable in terms of probability of adverse reaction/number of subjects/ year). Finally there is abiological cost associated with tolerability, a subjective parameter which is therefore difficult to quantify. In other words, even noninvasive modalities, like US or MR, may have a biological cost (small) related to risky manoeuvres or to the use of contrast media.

Availability is related to the diffusion of a given technology: it may be expressed in terms of average time to have access to a given technology and may vary widely, depending on the local situation.

Benefits are the final results we wish to obtain by using imaging modalities in terms of influence on the patients' final outcome. Naturally they are linked to

L. DALLA PALMA, C. RICCI and S. MAGNALDI cost; subsequently, no approach to diagnostic imaging modality can fail to consider the so-called cost-benefit ratio. A number of sub-categories of benefits can be singled out.

(1) Benefits affecting the quality of life: diagnostic information which leads to avoidance of further diagnostic procedures and/or useless therapies, enhances the quality of life.

(2) Benefits affecting therapeutic decisions means useful information for identifying the most appropriate therapeutic approach (medical, surgical, guided by diagnostic imaging itself)

(3) Purely diagnostic benefits are obtained when imaging modalities supply data for solving diagnostic problems generated by diagnostic imaging itself (especially when benign or malignant lesions are discovered incidentally).

The above factors should all be considered by the decision maker when drawing up the guidelines for optimal use of imaging protocols. A practical approach to the problem is examined for different clinical situations regarding diagnosis of the central nervous system (CNS) and abdominal field.

REFERRAL CRITERIA FOR SELECTION OF PATIENTS AND DIAGNOSTIC PROCEDURES IN CNS DIAGNOSTICS

As with US in abdominal diagnostics, the introduction of MRI as a neuroradiological tool has greatly reduced biological cost. Despite its high cost and its still limited availability, this modality is replacing CT more and more; coupled with high diagnostic accuracy, it uses safe contrast media and no ionising radiations.

Brain tumours

MRI is a superb method of studying brain tumours by virtue of its excellent contrast resolution, easy multiplanar imaging and absence of artefacts. MRI and CT are roughly equivalent for detection of most brain tumours but MRI is superior at the vertex, in the posterior fossa, near the wall of the middle fossa, at the base of the skull and in the orbit'3"5'. The criterion for selecting patients is based upon clinical features (accurate neurological examination, evaluation of electrical brain activity in basal conditions and after stimuli). In these cases the first imaging diagnostic step should be MRI despite its high financial cost since it is the most accurate and gives important benefits affecting therapeutic decisions. The only limiting factor is availability; thus, the criteria to select patients are the following:

Clinical features MRI (if available) EEG otherwise Brain evoked CT if negative MRI

potentials

Acute and chronic brain ischaemia Within a few hours of vascular occlusion, detection

and localisation of cerebral infarction is possible with MRI while CT often yields equivocal or negative results in the first 24 or 48 hours and MRI is still superior to CT in subacute and chronic stages'36-7'. As in brain tumours this would suggest employing MRI in all circumstances but MRI is less widely available than CT, especially for monitoring development of the disease. Furthermore, patients with severely impaired CNS function may require a complex anaesthesiological surveillance which is much more compatible with CT technology. The following criterion can be suggested:

Clinical MRI (if available) follow-up CT features otherwise

CT

CNS trauma It is important to remember that in clinical practice

patients undergo a plain film of the skull. This is for legal reasons although it has been shown that fractures of the skull not associated with specific clinical findings are very rarely responsible for brain lesions'8"10'. This shows how an apparently low cost modality has an unfavourable cost-benefits ratio in practice and should not even be considered for selecting patients to undergo more expensive modalities such as CT or MRI.

In head trauma, MRI has proved to be useful in detection of all types of intracranial haemorrhage, including haemorrhagic contusions and shearing injuries. In recent reports MRI shows a higher accuracy than CT. During the first one to three days after injury, however, CT may be preferable not only because it is better tolerated by patients but also because haemorrhage at this point may be more reliably demonstrated by CT'3"12'. The referral criteria are therefore the same as those suggested for ischaemic lesions.

REFERRAL CRITERIA FOR SELECTION OF PATIENTS AND DIAGNOSTIC PROCEDURES IN ABDOMINAL DIAGNOSTICS

Liver tumours

Primary liver tumours The role played by diagnostics and therapy of hepato

cellular carcinoma (HCC) has become increasingly important; in this field particularly diagnostic protocols must be carefully chosen in order to limit costs and enhance benefits. Patient recruitment must be clinical since categories of patients 'at risk' can be singled out

REFERRAL CRITERIA FOR PATIENT SELECTION on a clinical basis (patients with postnecrotic cirrhosis and/or abnormal serum level of alphafeto protein). The sensitivity of various imaging modalities"315' shows that a certain gain occurs when using modalities of a considerably higher cost (both financial and biological). For this reason, it is worthwhile employing US as a first diagnostic imaging step in patients recruited on a clinical basis. This allows us to create different categories of patients for whom different selection criteria must be considered in order to obtain the most favourable costbenefit ratio.

stop

In this situation US'reveals the presence of a large neoplasm (occupying more than 50% of the liver volume) and therefore the oncological stage overcomes the therapeutic possibilities regardless of the clinical stage of cirrhosis. The gain offered by this flowchart in reducing costs (financial and biological) as well as increasing benefits, in terms of quality of life is clearly evident, also when considering the complex protocol involving patients with early stage of the disease. (B) Clinical recruitment

(A) Clinical recruitment

US (+biopsy)

extensive involvement

US (+biopsy) DSA (digital subtraction angiography) LCT (lipiodol computed tomography)

\

early stage early stage

CT early stage

early stage

advanced stage In this second situation, an early stage of the focal disease at US examination associated with an early stage of the diffuse disease (cirrhosis), it is important to define the actual early stage of the HCC as accurately as possible. This is usually achieved by employing sequentially three modalities which, at increasing costs, ensure high sensitivity and therefore enable the most appropriate treatment (surgical or nonsurgical) to be set. The increased costs (both financial and biological) are offset by improved treatment and good results. (C) Clinical US (+biopsy) locally CT DSA recruitment advanced

US and confirmed by CT, DSA and LCT, and/or associated with general conditions which preclude surgical treatment, patients can be selected for CEAT or PEI as an alternative, at lower biological cost.

Secondary liver tumours As in primary tumours, patients at risk for metastatic

disease should be first selected on the basis of clinical criteria. Diagnostic efforts should be mainly concentrated on patients who have a primary tumour which is well treatable and possibly curable and in whom treatment of an early stage of metastatic disease can significantly change the medium to longterm survival rate; for example, metastases from breast, colon and kidney.

The sensitivity of diagnostic imaging is related to lesions size"6"19' and, although high levels of sensitivity are reached by more expensive modalities like CT or MRI, US remains, by virtue of its low cost and wide availability, the first tool in diagnostic protocols. A suggested criterion can therefore be the following; Clinical US * multiple lesions Stop recruitment \ single lesion CT

single lesion > US or CT guided biopsy

Liver haemangiomas The most common benign focal lesions of the liver

are generally asymptomatic. They are accidentally discovered in patients undergoing US examination for a variety of reasons. When they show a typical US pattern there is no diagnostic problem but when the pattern is atypical and in patients likely to have a metastatic disease they deserve a more careful evaluation. US, CT and DSA have a comparably high sensitivity but MRI and nuclear medicine (NM) are definitely more sensitive'202".

Therefore, the use has been suggested, in ambiguous cases, an association of these modalities'22' by applying the following criterion: US incidental discovery

ambiguous pattern or at risk patients

/ \

CT MRI (if available) NM

/ no treatment LCT * CEAT (chemo embolisation

arterial transcatheter) \ PEI (percutaneous ethanol

injection) In this situation, a locally advanced disease revealed by

This criterion bears out the idea that modalities which have high financial cost but low biological cost (MRI and NM) can be used to solve problems created by other imaging modalities avoiding more invasive and costly steps (DSA).

Jaundice When the clinical onset and laboratory tests are

ambiguous in determining the aetiology of the jaundice, US has been seen to be very accurate in differentiating obstructive from nonobstructive forms. Unfortunately,

L DALLA PALMA, C. RICCI and S. MAGNALDI for anatomical and physical reasons, its accuracy is not equally high in detecting the site and nature of the obstruction. US features are nevertheless useful in selecting patients who must undergo more expensive and invasive procedures like CT or percutaneous transhepatic cholangiography (PTC)'23-28'.

The gain, in terms of financial and biological cost, is evident when using the following criterion:

Clinical US non-obstructive stop feature \ obstructive Laboratory > site/nature undetected

CT or PC or ERCP (if therapeutic approach)

Cholelithiasis

Until ten years ago this disease, affecting a large population in the west, was diagnosed by means of oral cholecystography (OC) which uses both ionising radiation and contrast media but whose accuracy average is lower in comparison with US'29-35'. Moreover it is not able to depict the gallbladder wall and is rather time consuming when a functional evaluation of the gallbladder wall is required. For these reasons, as it is more sensitive and at least as specific as OC, US has now replaced this modality. However, if the clinical feature suggests that a stone or stones have moved to the choledochal duct (recent biliary colic with onset of jaundice) US accuracy decreases'36' for the above mentioned reasons (see above protocol for jaundice) and a different approach should be considered.

Clinical feature US (pain, dyspepsia)

Clinical feature see protocol for (biliary colic, jaundice, fever) jaundice

Kidney

Renal tumours

Although accuracy of conventional intra venous urography (IVU), in expert hands, may be rather high, new modalities have dramatically increased the possibilities of detecting the disease in early stages'37-4"; US in particular, being a very diffuse modality, often allows recognition of this pathology when it is still asymptomatic. As in liver metastases, accuracy is related to tumour size and to its echo pattern. However CT is most accurate and must be used for integrating US data and for staging. IVU plays a minor role and this once again leads to a reduction in biological cost for the whole population since more costly modalities are dedicated to fewer well defined cases in which the US pattern is suspicious or atypical. The criterion for selecting patients and procedures is therefore:

Clinical feature

US positive or suspicious

CT (+ staging)

Reno-ureteral colic

In this clinical situation IVU remains a reference tool by virtue of its superior spatial resolution. Both the excretory pathway and small stones are usually depicted with great accuracy. However, it has recently been proven that a plain film (PF) of the abdomen (which requires a small fraction of IVU exposure to X rays) combined with US examination of kidneys and distal ureters is almost as accurate as IVU'4243'. This represents a considerable gain in biological cost and enables us to select a limited number of patients for IVU, avoiding exposure of the genital area in sometimes very young patients.

Clinical PF + US negative follow-up or IVU feature

Prostatic carcinoma

There is increasing interest in this pathology due to the fact that it may be cured when it is still confined within the gland borders (B or T2 stage). Once the disease has been detected by the urologist and histologically proven it is not useful to employ ionising radiations to stage it since CT has a rather poor accuracy compared to MRI'44"47'. In this case, however, a second clinical-epidemiological criterion must be applied in order to obtain maximum benefit from staging; only subjects with life expectation of at least ten years and in good general condition should undergo staging with MRI. The composite criterion will therefore be:

Clinical Histological (US if clinical MRI feature guided) diagnosis- condition Laboratory grading satisfied

CONCLUSIONS

A few examples show that criteria for selecting patients and imaging modalities must be adapted to each clinical situation. The 'first priority' level is always based upon clinical data, epidemiological data, and laboratory tests. Diagnostic modalities are chosen on the basis of composite considerations in which the cost-benefits ratio must play the main role. In other words, this implies a priori knowledge of the efficacy of all modalities in each clinical situation and of benefits to be obtained. A clear trend towards a reduction in biological cost is evident in modern diagnostic imaging protocols.

REFERRAL CRITERIA FOR PATIENT SELECTION REFERENCES

1. Lucertini, M. and Telmon, D. Benchmarking: valuatazione e miglioramento delle prestazioni aziendali (in press). 2. Inouye, S. K. and Sox, H. C. Standard and Computed Tomography in the Evaluation of Neoplasms of the Chest: a Compara

tive Efficacy Assessment. Ann. Int. Med. 105, 906924 (1986). 3. Orrison, W. W., Stimac, G. K., Stevens, . ., Lamasters, D. L., Espinosa, M. C , Cobb, L. and Mettler, F. A. Comparison

of CT, LowField Strength MR Imaging, and HighField Strength MR Imaging. Radiology 181, 121127 (1991). 4. Schoerner, W., Schubeus, P., Henkes, H., Rottacker, C, Hamm, . and Felix, R. intracranial Meningiomas; Comparison

of Plain and Contrastenhanced Examinations in CT and MRI. Neuroradiology 32, 1218 (1990). 5. Tampieri, D., Moumdjian, R., Melanson, D. and Ethier, R. Intracerebral Gangliogliomas in Patients with Partial Complex

Seizures; CT and MR Imaging Findings. Am. J. Neuroradiol. 12, 749755 (1991). 6. Bryan, R. N., Levy, L. M., Whitlow, W. D., Killian, J. M., Preziosi, T. J. and Rosario, J. A. Diagnosis of Acute Cerebral

Infarction: Comparison of CT and MR Imaging. Am. J. Roentgen. 157, 585594 (1991). 7. Seiderer, M., Krappel, W., Moser, E., Hann, D., Schmiedek, P., Buell, V., Kirsch, C. M. and Lissner, J. Detection and

Quantification of Chronic Cerebrovascular Disease: Comparison of MR Imaging, SPECT, and CT. Radiology 170, 545548 (1989).

8. Bell, R. and Loop, J. The Utility and Futility of Radiographic Skull Examination for Trauma. New Engl. J. Med. 284, 236239 (1971).

9. Masters, S. J. Evaluation of Head Trauma: Efficacy of Skull Films. Am. J. Roentgenol. 135, 539547 (1980). 10. National Study by the Royal College of Radiologists. Cost and Benefits of Skull Radiography for Head Injury. Lancet 2,

791795 (1981). 11. Gentry, L. R., Godersky, J. C , Thompson, B. and Dunn, V. D. Prospective Comparative Study of Intermediatefield MR

and CT in the Evaluation of Closed Head Trauma. Am. J. Neuroradiol. 9, 91100 (1988). 12. Hesselink, J. R., Dowd, C. F., Healy, M. E., Hajek, P., Baker, L. L. and Luerssen, T. G. MR Imaging of Brain Contusions:

A Comparative Study with CT. Am. J. Neuroradiol 9, 269278 (1988). 13. Choi, B. I., Park, J. H., Kim, B. H., Kim, S. H., Han, M. C. and Kim, C. W. Small Hepatocellular Carcinoma: Detection

with Sonography, Computed Tomography (CT), Angiography and LipiodolCT. Br. J. Radiol. 62, 897903 (1989). 14. Takayasu, K., Moriyama, ., Muramatsu, Y., Makuuchi, M., Hasegawa, H., Okazaki, N. and Hirohashi, S. 77ie Diagnosis

of Small Hepatocellular Carcinomas; Efficacy of Various Imaging Procedures in 100 Patients. Am. J. Roentgenol. 155, 4954 (1990).

15. Takashima, T., Matsui, O., Suzuki, M. and Ida, S., Diagnosis and Screening of Small Hepatocellular Carcinomas. Radiology 145, 635638 (1982).

16. Matsui, O., Takashima, T., Kadoya, M., Suzuki, M., Hirose, J., Kameyama, T., Choto, S., Konishi, H., Ida, M. and Yamaguchi, A. Liver Metastases from Colorectal Cancers: Detection with CT during Arterial Portography. Radiology 165, 6569 (1987).

17. Stark, D. D., Wittenberg, J., Butch, R. J. and Ferrucci, J. T. Hepatic Metastases: Randomized, Controlled Comparison of Detection with MR Imaging and CT. Radiology 165, 399^*06 (1987).

18. Vlachos, L., Trakadas, S., Gouliamos, ., Lazarou, S., Mourikis, D., Ioannou, R., Kalovidouris, A. and Papavasiliou, C. Comparative Study Between Ultrasound, Computed Tomography, IntraArterial Digital Subtraction Angiography, and Magnetic Resonance Imaging in the Differentiation of Tumors of the Liver. Gastrointest. Radiol. 15, 102106 (1990).

19. Wemecke, K., Rummeny, E., Bongartz, G., Vassallo, P., Kivelitz, D., Wiesmann, W., Peters, P. E., Reers, ., Rieser, M. and Pircher, W. Detection of Hepatic Masses in Patients with Carcinoma: Comparative Sensitivities of Sonography, CT, and MR Imaging. Am. J. Roentgenol. 157, 731739 (1991).

20. Stark, D. D., Felder, R. C. and Wittenberg, J. Magnetic Resonance Imaging of Cavernous Hemangioma of the Liver; Tissue Specific Characterization. Am. J. Roentgenol. 145, 213220 (1985).

21. Engel, . ., Marks, D. S., Sandler, M. A. and Shetty, P. Differentiation of Intrahepatic Lesions with 99m TcRed Blood Cell Imaging. Radiology 146, 777782, 1983.

22. Nelson, R. C. and Chezmar, J. L. Diagnostic Approach to Hepatic Hemangiomas. Radiology 176, 1113 (1990). 23. Baron, R. L., Stanley, R. J., Lee, J. K. T., Moehler, R., Melson, L., Balfe, D. M. and Weyman, P. J. A Prospective Comparison

of the Evaluation of Biliary Obstruction using Computed Tomography and Ultrasonography. Radiology 145, 9198 (1982). 24. Dalla Palma, L., Maffessanti, M., Bazzocchi, M., Rizzatto, G., PozziMucelli, R. S. and Brizzi, F. Combined Use of Ultrasono

graphy and Transhepatic Percutaneous Colangiography in Obstructive Jaundice. Eur. J. Radiol. 2, 281289 (1982). 25. Dalla Palma, L., Rizzatto, G., Bazzocchi, M., PozziMucelli, R. S., Brizzi, F. and Maffessanti, M. L'ecografia nella valuta

zione del paziente itterico: attendibilit su 246 pazienti controllati. Radiol. Med. 67, 615622 (1981). 26. Gibson, R. N., Yeung, E. and Thompson, J. N. Bile Duct Obstruction: Radiologie Evaluation of Level, Cause and Tumor

Resectability. Radiology 160, 43^*7 (1986). 27. Pedrosa, C. S., Casanova, R Lezana, A. H. and Fernandez, M. C. Computed Tomography in Obstructive Jaundice. Part

II: The Cause of Obstruction. Radiology 139, 635645 (1981). 28. Pedrosa, C. S Casanova, R. and Rodriguez, R. Computed Tomography in Obstructive Jaundice. Part I: The Level of Obstruc

tion. Radiology 139, 627634 (1981).

L DALLA PALMA, C. RICCI and S. MAGNALDI 29. Coopenberg, P. L. and Burhenne, M. Z. Real Time Ultrasonography: Diagnostic Technique of Choice in Calcious Gallbladder

Disease. New Eng. J. Med. 302, 12771279 (1980). 30. Crade, M., Taylor, K. J. W., Rosenfield, A. T., Capper, S., DeGraaf, F. and Minihan, P. Surgical and Pathological Correlation

of Cholescistosonography and Cholecistography. Am. J. Roentgenol. 131, 227229 (1978). 31. De Lacey, G., Gajjar, B. and Towomey, B. T. Should Colecistography or Ultrasound be the Primary Investigation for

Gallbladder Disease? Lancet i, 205207 (1980). 32. Klingensmith, W. C. and Eckhout, G. V. Cholelithiasis in the Morbidly Obese: Diagnosis by US and Oral Cholecystography.

Radiology 160, 2728 (1986). 33. Krook, P. M., Allen, F. M., Bush, W. M., Maimer, G. and MacLean, M. D. Comparison of the Real Time Cholecystosonogra

phy and Oral Cholecystography. Radiology 135, 145148 (1980). 34. Wolson, A. M. and Goldberg, B. B. Gray Scale Ultrasonic Cholecystography, a Primary Screening Procedure. J. Am. Med.

Assoc. 240, 20732075 (1978). 35. Mcintosh, D. M. F. and Pennali, M. F. Gray Scale Ultrasonography as a Screening Procedure in the Detection of Gallbladder

Disease. Radiology 136, 725727 (1980). 36. Mitchell, S. E. and Clark, R. A. A Comparison of Computed Tomography and Sonography in Choledocolithiasis. Am. J.

Roentgenol. 142, 729733 (1984). 37. Amendola, . ., Bree, . L., Pollack, H. M., Francis, I. R., Glazer, G. M., Jafri, S. Z. H. and Tomaszewski, J. E. Small

Renal Cell Carcinoma: Resolving a Diagnostic Dilemma. Radiology 166, 637664 (1988). 38. Bazzocchi, M., PozziMucelli, R. S, Zanella, F. and Ricci, C. Diagnostica 1985 delle neoplasie renali. In: Radiourologia

1985, pp. 8194 (Lint, Trieste) (1985). 39. Mobilio G., Cavalli, A. and Bianchi, G. Choice between US and CT in the Radiological Approach to the Renal Masses.

Diagn. Imaging 51, 3238 (1982). 40. Plainfoss, M. C , Delecoeullerie, G. and Vital, J. L. 165 Renal Carcinomas, Accuracy of Imaging for Diagnosis and Spread

cost Efficiency. Eur. J. Radiol. 3, 132137 (1983). 41. Smith, S. J., Bosniak, . ., Megibow, A. J., Hulnick, D. H., Horii, S. C. and NageshRaghavendra, B. Renal Cell Carcinoma:

Earlier Discovery and Increased Detection. Radiology 170, 699703 (1989). 42. Bazzocchi, M., Stacul, F., Cressa, C. and Dalla Palma, L. L'ecografia nella colica renale. Radiol. Med. 76, 7882 (1988). 43. Dalla Palma, L., Stacul, F., Bazzocchi, M., Pagnan, L., Festini, G. and Marega, D. Ultrasonography and Plain Film versus

Intravenous Urography in Ureteric Colic. Clin. Radiol. 47, 333335 (1993). 44. Biondetti, P. R., Lee, J. K. T., Ling, D. and Catalona, W. J. Clinical Stage Prostate Carcinoma: Staging with MR Imaging.

Radiology 162, 325329 (1987). 45. Emory, T. H., Reinke, D. B., Hill, A. L. and Lange, P. H. Use of CT to Reduce Understaging on Prostatic Cancer: Compari

son with Conventional Staging Techniques. Am. J. Roentgenol. 141, 351354 (1983). 46. Morgan, C. L., Calkins, R. F. and Cavalcanti, E. J. Computed Tomography in the Evaluation, Staging and Therapy of

Carcinoma of the Bladder and Prostate. Radiology 140, 751761 (1981). 47. Pontes, E., Eisenkraft, S., Watanabe, H., Ome, H., Saiton, M. and Murphy, G. P. Preoperative Evaluation of Localized

Prostatic Carcinoma by Transrectal Ultrasonography. J. Urol. 134, 289291 (1985).

Radiation Protection Dosimetry Vol. 57, Nos. 1-4, pp. 9-11 (1995) Nuclear Technology Publishing

THE SCIENTIFIC WORK-UP OF RADIOGRAPHIC IMAGE QUALITY NOW AND A DECADE AGO: THE RADIOLOGIST'S APPROACH J. W. Oestmann Diagnostische Radiologie I Klinikum der Georg-August-Universitt Robert-Koch-Strae 40 Gttingen, Germany

INVITED PAPER

Abstract The past decade has seen a number of efforts to increase image quality and decrease examination hazards to the patient in medical imaging. A comparison of image quality studies in the early 1980s with those of the early 1990s shows that the amount of scientific publications on the topic as well as their sophistication has grown. The concept of diagnostically relevant structures as well as the observer performance test with receiver operator characteristics (ROC) analysis has been well received. On the basis of our own experience in projection radiography a three-step approach to the evaluation or optimisation of new imaging modalities is proposed: (1) base evaluation applying adequate physical parameters; (2) subjective optimisation of the system with the diagnostic objectives in mind (i.e. the diagnostically relevant structures in radiographic studies) to generate a set of alternative hypotheses; and (3) comparison of these hypotheses with clinically oriented observer performance tests.

INTRODUCTION

The past decade has seen a number of scientific and political efforts to increase image quality and decrease examination hazards to the patient in medical imaging. On the political side, national radiation protection laws and international and EEC initiatives led to the development and further improvement of technical standards and to the formation of local authorities which survey the quality of radiographic imaging with special attention to dose and diagnostic quality. On the scientific side, the advent of new modalities spawned interest in methodologies which permit a reliable and reproducible determination of diagnostic value. In conventional radiographic technique, dose reducing combinations of films with rare earth screens and high and low latitude films had to be tested. The introduction of commercial digital radiography systems in the early 1980s made imaging scientists search for subtle yet relevant differences between digital and conventional images. The post-processing capabilities of these systems could only be optimised with sophisticated tools. The rapid development and spread of CT, ultrasound and MRI required scientifically based decisions on the diagnostic value of each modality. The rise and fall of thermography was seen as an expensive example of a failed scientific evaluation. The importance of these efforts was enhanced by increasing cost constraints in national health programmes and the international economy.

This paper is intended to describe briefly how image quality has been described scientifically and how we think image quality should be evaluated. An analysis of

scientific publications of the past decade will show how interest and sophistication of methodology have improved.

THE RADIOLOGIST'S DEFINITION OF IMAGE QUALITY

The radiologist's view of image quality is dominated by his or her professional task: the detection or exclusion of pathology as well as the differential diagnosis and localisation of detected abnormalities. A good detectability of disease is achieved if the conspicuity of a lesion is high, i.e. if the perceptual difference between the lesion and the anatomical background is large"-2). If the characteristics of the abnormality are quite different from normal anatomical structures (such as an enhancing neurofibroma in MRI), the documentation of normal anatomical structures is of lesser importance for the detection task. If the abnormality is, however, similar in imaging characteristics to physiological structures (such as pulmonary nodules and pulmonary vessel intersections), these structures must also be optimally imaged. This is of particular importance for the differential diagnosis and localisation of lesions. The image quality of a given modality is thus defined by its ability to demonstrate disease with a high level of conspicuity and to delineate anatomical structures relevant for detection, differential diagnosis and localisation. Radiologic image quality in our view is always a relative parameter. It defines itself in comparisons of different imaging modalities (such as CT and MRI) or different protocols of one imaging modality (such as different

J. W. OESTMANN dose levels or postprocessing algorithms in digital radiography (DR)).

EVALUATION OF IMAGE QUALITY IN THE LAST DECADE

An analysis of the medical literature over the past decade with the help of a computer database literature search using the keywords 'image quality' and Observer performance' shows that the number of publications on image quality issues has grown continuously (Figure 1 ). A separate analysis of papers published in one major English language (Radiology, sample size: 330 papers) and one major German language (Rfo, sample size: 182 papers) radiological journal in 1982 and 1992 supports this statement: In Radiology the percentage of papers on the topic rose from approx. 12 to 17% while this percentage grew from approx. 6 to 24% in Rfo. There seems to be a certain lag time between publications on the topic in the English language and the German language journal. Such papers were considered for further study that made a statement on the comparative image quality of imaging modalities as defined above. The analysis according to imaging modalities for both journals (Figures 2 and 3) shows that filmscreen systems

200

Q. (0

100 -Q E

82 83 84 85 86 87 88 89 90 91 92 93 (18) Year

Figure 1. Scientific publications on image quality in the medical literature between 1982 and 1992. For 1993, only the first

eight months could be analysed.

Total FS DR CT US/ MR Angio. Nucl. Dop.

(FSS) and computed tomography (CT) were in the centre of scientific interest in the early 1980s. In the early 1990s magnetic resonance imaging (MRI) and digital radiography (DR) were the subjects of a large number of studies. Partially due to new developments in CT (spiral CT) and in conventional radiography (AMBER) but also as part of comparative studies including MR and DR, research on CT and FFS continued at a high level. For ultrasound and Doppler techniques combined, interest remained similar. Papers also were classified according to methodology following these definitions: (1) Undefined or n