Soft Copy versus Hard Copy Reading in Digital Mammography

Abstract

The objective of this study was to compare soft copy reading at a mammography work station with hard copy reading of full-field digital mammographic images. Mammograms of 60 patients ( n = 29 malignant, n = 31 benign) performed with full-field digital mammography (Senographe 2000D, GE, Buc, France) were evaluated. Reading was performed based on hard copy prints (Scopix, Agfa, Leverkusen, Germany) and on 2 k x 2.5 k high-resolution monitors (Sun Ultra 60, Sun Microsystems, Palo Alto, California, USA). Four readers with different levels of experience in mammography categorized the mammograms according to the BI-RADS classification. The comparative study was performed by four readers, and at least 2 months elapsed between the reading sessions. Postprocessing, of course, was available only at the work station (windowing and leveling, zooming, inversion). Sensitivity, specificity, and positive predictive value were evaluated. Diagnostic accuracy of the evaluation was determined. Sensitivity for malignant lesions in hard copy versus soft copy reading was 97% vs 90%, 97% vs 97%, 93% vs 97%, and 76% vs 76% for the four readers, respectively. Specificity was 52% vs 68%, 58% vs 74%, 65% vs 48%, and 61% vs 68%. Accuracy for the classification of malignant lesions according to the BI-RADS categories showed no difference between hard copy and soft copy reading. Soft copy reading is possible with the available system and enables radiologists to use the advantages of a digital system.

Diagnostic performance of detecting breast cancer on computed radiographic (CR) mammograms: comparison of hard copy film, 3-megapixel liquid-crystal-display (LCD) monitor and 5-megapixel LCD monitor

“…Some studies have evaluated the efficacy of soft-copy reading of digital mammography using a flat panel detector. Two studies have demonstrated that soft-copy reading using 5-MP CRT monitors is as accurate as hardcopy reading [19,20]. In addition, 5-MP CRT monitors and 5-MP LCDs were revealed to be equivalent for evaluating the normal anatomy and features of abnormalities, simulated masses, and microcalcifications [21,22].…”

Characterization of microcalcification: can digital monitor zooming replace magnification mammography in full-field digital mammography?

“…Several investigations have shown that interpretation with soft-copy display is likely to be useful with digital mammography and is unlikely to change accuracy or speed compared with interpretation using hard-copy display of digital mammography [6,11]. This is the case despite the fact that the spatial resolution of a workstation monitor is lower than that of film for printing [12][13][14].…”

“…They strongly recommended the post-processing of images, including adjustments of window level and zoom, during soft-copy interpretation [4,5]. Obenauer et al reported that further studies of tools in soft-copy reading are needed to realize its potential benefits [6].…”

Evaluation of Low-Cost Telemammography Screening Configurations: A Comparison with Film-Screen Readings in Vulnerable Areas

“…For rural or underserved areas, telemedicine may provide a cost-effective solution for screening mammography programs and computed radiography (CR) and full-field digital mammography (FFDM) are useful in the implementation of telemammography services. Although previous studies have reported no significant differences between film-screen mammography and digital modalities [6][7][8], this latter technology may be unaffordable for vulnerable populations, where only conventional film-screen mammograms may be available. Therefore, the digitization of films is required to provide telemammography services.…”