An overview of the UK breast screening program is presented and the importance of observer performance to its success is shown. A conceptual model is presented which leads to the consideration of diagnostic errors in three classes. Appropriate training can help reduce the occurrence of these errors. Data are presented from a national self assessment program which aims to give radiologists insight into aspects of their performance. In particular, information on calcification detection and the benefits of double reading is considered as a means of improving cancer detection rates.

emerged from these schemes probably mainly due to difficulties in case follow-up. This is an inevitable consequence when women with a suspected abnormal mammogram are referred elsewhere for assessment. Some of the schemes were well controlled, high-quality services, but others were far from this, with abysmal standards of mammography, and mammographic interpretation which was in some instances so poor as to be positively dangerous to the users of the schemes, with high numbers of false positives and, with possibly even worse effects, false negatives. One suspects that financial reward was a major objective of some of the organisations and individuals who initiated some of these early breast screening schemes. The NHS activity at this stage was limited to a few centres undertaking mammography of symptomatic women with only a handful of these involved with trials of breast screening. A multi-centre UK trial of early detection had been in progress since 1979, designed to test the relative values of various screening methods,’ but a report on the preliminary findings was not due until 1988. The public, lead by women’s organisations and supported by a sprinkling of enthusiastic professionals, was beginning to press for breast cancer screening as a part of the NHS. The health ministers, however, were rightly cautious. They, perhaps more than anyone, were aware of the less than optimal service which had resulted from the over-hasty, largely uncontrolled development of the Cervical Cancer Screening Programme, initiated on a wave of political enthusiasm as a result of women’s pressure groups.

The early experience of the Nottingham Breast Screening Service participating in the prevalent round of the UK National Health Service Breast Screening Programme is presented. Over a 3-year-period management protocols have been developed and refined. These protocols are based on the principle of multidisciplinary triple assessment, including imaging, clinical evaluation and fine needle aspiration cytology (FNAC). Throughout this period screening performance indicators well within the national guidelines were achieved. In addition, recall for assessment rates fell from 12% to 3.4%, the benign-to-malignant biopsy ratio falling from 1:1 to 0.3:1, while the cancer detection rate of 7 per 1000 women screened was maintained. A progressive increase in diagnostic specificity was achieved while retaining cancer detection sensitivity.

For breast cancer screening by mammography to be effective, the processes involved must operate to the highest possible standards if the ultimate goal of reducing the mortality from breast cancer - while causing the minimum of morbidity - is to be achieved. Before embarking on any breast screening programme, the educational needs of the health-care teams to be involved must be clearly understood. It must be recognised from the outset that the mammographie screening of asymptomatic women is entirely different from the diagnostic processes involved in the assessment of symptomatic breast disease. It is very unlikely that the required expertise to undertake breast cancer screening and the management of screen-detected lesions will exist in sufficient depth and breadth within a health-care system which has not previously been involved in breast screening. It is a prerequisite of any screening programme that a comprehensive teaching programme be instituted for all the healthcare professionals who will be involved in the screening process. It must also be recognised that these training programmes must be set up to provide continuing education so as to ensure that those involved are kept fully informed of all the relevant advances and changes in the field. This chapter provides a brief outline of the educational requirements of any population-based mammographie screening programme.

A key issues in the introduction and development of appropriate image processing techniques in mammography is the establishment of the current performance of radiologists in the area. This is necessary because the utility of machine vision approaches is largely validated by comparison with known radiological performance measures (which may well be variable) on the same set of cases. Furthermore the determination of weaknesses in existing human mammographic interpretative ability will demonstrate where machine vision approaches are currently most needed and thus likely to be of maximum benefit in a breast screening program. Following the introduction of breast screening in the U.K. a national self-assessment program has been implemented for all radiologists involved in this specialty. One of the outcomes of this program is the determination of radiological performance variations on this standard task. It is argued that these demonstrate the need for any machine vision approach to take such individuality into account before it can be implemented usefully.

Since 1986 national breast screening has been introduced throughout the UK. This involves all women in the age range 50–64 years and generates some 1.5 million mammograms annually. Currently specialised radiologists visually examine each case for the presence of cancer; the reported detected incidence of which is 6.2 cases per 1000 women examined. The potential for computer aided diagnosis lies in the difficulty of the task for human observers to detect this disease at an early stage of presentation from amongst the very large number mammograms of normal women examined. Other existing research programmes are currently investigating machine detection of particular mammographic features, such as microcalcifications, and the performance of these approaches typically is compared to that of expert radiologists. We present data from the first two years of a national programme for screening radiologists in the UK. One outcome of which is the demonstration of the typical performance characteristics of radiologists so illustrating where computer aided diagnostic systems can contribute most as an aid to radiologists in breast screening.

To evaluate the place of mammography in the selection of patients for excision and radiotherapy for primary breast cancer a detailed analysis of pre-operative mammograms was performed in (i) a study group of 37 patients who developed local recurrence; (ii) a matched control group with a median local recurrence free survival of 57 months. There were significantly more multifocal tumours in the study group. Tumours were significantly larger (P = 0.02) and closer to the nipple (P = 0.008) in the study group compared to the control group. Regular follow-up mammograms were available in 26 of the study group. Twenty-one patients had mammographic evidence of either residual or recurrent tumour. We conclude that pre-operative mammography is essential in the selection of patients for excision and radiotherapy. Following treatment, mammography is useful in detecting residual or recurrent disease.