Digital mammography.

IN 1992, A NATIONAL Cancer Institute expert panel determined that, of all emergent technologies, full-field digital mammography (FFDM) held the greatest potential for improving breast cancer detection.1,2 The transition to digital mammography is a necessary step in the ongoing process of improving mammographic image quality and advancing toward the goal of earlier breast cancer detection. Digital mammography will also enable the completion of the conversion of radiology departments to fully digital operation. The potential of digital imaging in the field of diagnostic and screening mammography is made clear in an evaluation of conventional screen-film mammography. Screen-film mammography is widely accepted as an effective breast cancer screening modality. Large randomized screening trials have shown that routine screen-film mammography reduces the breast cancer mortality rate in women age 50 years and older by up to 30% when compared with unscreened controls.3-5 Moreover, cancers found by screening mammography tend to be smaller and less advanced than those found by breast physical examination, and patients with early stage breast cancer tend to have better survival rates.6,7 The success of screen-film mammography is based on the high spatial resolution and the low contrast sensitivity achieved through improvements in x-ray tube design, screen-film combinations, grids, and film processing. The result has been better image quality, lower patient dose, and most importantly, the ability to detect small, nonpalpable breast cancers. However, despite these advances, film-screen mammography operates with a sensitivity rate for women with dense breasts of less than 70%.8 The relatively poor sensitivity in dense glandular women is in part because of insufficient display contrast between the potentially malignant lesion and the surrounding benign glandular breast tissue. This lack of display contrast is because of the limited range of exposures over which screen-film systems can display subtle contrast differences. Because of the shape of the screen-film exposure response curve, the range of x-ray exposures and optical densities displayed on any single film (called the latitude of the film) is quite limited (Fig 1). When a lesion is detected mammographically, the specificity, or the ability of the radiologist to predict whether the lesion represents a malignancy, is unfortunately quite low. The additional workup necessary to characterize a mammographically detected lesion may include multiple additional images such as spot compression, magnification views, or images using different exposures. Frequently, these additional images are necessary because of the technical limitations of the screenfilm system. Unfortunately, only 5% to 40% of mammographically detected breast lesions prove to be malignant at biopsy. Digital mammography is believed to have the potential to be more sensitive and specific than screen-film mammography. The sensitivity may be improved by the wide dynamic range of the digital detector, which allows improved detection of lesions in all areas of the breast with a single exposure. The specificity may be improved by the ability to manipulate the image data after acquisition to optimize detection in all breast types (Fig 2). Because digital systems acquire, display, and store the image data independently, each of these functions can be individually optimized. A comparison of these and other characteristics of filmscreen and digital mammography is provided in Table 1.