A comparison of four mammographic image quality test objects

Abstract

The need for rigorous image quality assurance in mammography is well documented (Pritchard Report, 1989). However, there is still debate over the suitability of those image quality test objects commercially available. We have been able to compare three contemporary test objects with the “Barts” test object which has been in use since 1980 and has been used as a standard by many workers (Kirkpatrick & Law, 1987). The four mammographic test objects have been compared using six mammographic film-screen combinations. The following test objects were used. The “Barts” mammographic image quality test object (White & Tucker, 1980). This object contains several series of opacities designed to mimic mammographic diagnostic features, some being fat equivalent, some water equivalent, some skin equivalent and some equivalent to calcifications, which are immersed in an epoxyresin scattering medium. The object is designed to assess contrast and resolution combined in various ways. The Dupont mammographic image quality test object (Ackermann, personal communication, 1988). This simulates microcalcifications, fibrous structures and small spherical objects. It also allows an assessment of minimal detail contrast, line resolution in two dimensions and point resolution. A selection of organic materials is included to simulate benign and malignant tumours, lymph nodes and tissues characteristic of a dense breast. It is designed to be used with a stack of attenuator plates to provide the appropriate absorption, scatter and geometrical unsharpness in test exposures.

The Design and Fabrication of a Full Field Quantitative Mammographic Phantom

“…In order to monitor this high image quality and to show that it has not deteriorated over a long period of time, it usual to use one or more image quality phantoms and to take radiographs of these at regular intervals [1] . The need for rigorous image quality in mammography is well documented [2] . Appropriate phantoms are needed to assess image quality, equipment performance and patient exposure.…”

“…As currently DM systems are becoming a more widely used clinical tool, new methodologies for QA are required to ensure that they produce an image containing all the information essential for diagnosis. Although the need for image QA in DM is well documented there is still debate over the suitability of the image quality test objects (phantoms) that are commercially available [2] . Presently available phantoms are not adequate for DM which need to be fast, easy to use, provide the user with a clear picture of high and low contrast resolution over the full field of view and demonstrate geometrical distortions in the vertical, horizontal or diagonal directions.…”

“…The shape and density of the test objects must be representative of those present in breast tissue in order to make a fair comparison, and finally DM QC phantom images should be able to be automatically analysed [6] . The lack of a suitable QC phantom [2,3] as well as the improvement of contrast and image quality in DM requires a newly designed test object (phantom), which is matched to the particular characteristics of digital imaging systems. Implementation of DM is a more costly procedure and perhaps unobtainable for most mammography practices in the short term, but it will occur in future [7] .…”

Design Biological Phantom for Human Breast

“…1 The need for accurate image quality in mammography is well documented. 2 Imaging of test phantoms provides information about the performance of the system in terms of measurable physical parameters such as high and low contrast resolution and enables the comparison of different imaging systems. Any given phantom though may not be well suited to both of these objectives because of their different requirement for complexity and ease of interpretation.…”

Evaluation methods of mammographic imaging

“…The design of these phantoms depends on the goals of evaluation studies, whether the performance of selected parts or of the entire imaging chain is to be checked, or absolute or relative values are to be derived. These phantoms contain different types of test objects; typically include a step wedge, high contrast small-size circular objects, high contrast irregular objects simulating calcifications, fibres (curvilinear structures), low contrast circles representing lesions, more extensive contrast-detail objects and resolution gratings [18][19][20][50][51][52][53][54][55][56][57][58][59], In table 2.1 the essential features of mammographie physical phantoms are provided.…”