Effects of the DICOM grayscale standard display function on the accuracy of medical-grade grayscale and consumer-grade color displays for telemammography screening

Abstract

The aim of this study was to compare the diagnostic accuracy of consumer-grade and medical-grade monitors with very different costs for breast cancer detection when used with and without the Digital Imaging and Communication in Medicine (DICOM) standard calibration. This study was retrospective, with a factorial design and repeated measures, and used 70 digital mammograms (40 benign or normal cases and 30 malignant cases), four radiologists, and three displays with and without display calibration. Film mammograms were included. The readings were classified according to the Breast Imaging Reporting and Data System. One medical-grade grayscale display and two consumer-grade displays were compared. Receiver operating characteristic curves were plotted for nodules, micro calcifications, and the degree of malignancy. The diagnostic accuracy of each device was calculated as the area under the curve, and the accuracy was compared using an analysis of variance. INTRODUCTIONBreast cancer is globally one of the most prevalent diseases among women over 50 years of age and presents the highest rate of cancer mortality in women in almost all countries. To detect this disease in its early stages, screening programs have been developed that include, among other methods, routine annual or biannual mammograms. 1-3 Screening mammography has shown effectiveness in reducing mortality, which is between 9% and 32% in women aged 40 years or older. 4, 5 Telemammography may be a solution to expand the screening coverage in a timely and efficient manner in vulnerable populations. This screening technique requires the use of digital images, but the cost of specialized equipment to generate, digitize and view digital mammograms is very high, resulting in the need for a low-cost alternative in developing countries. Before consumer-grade devices are adopted in the medical field, their diagnostic accuracy should be determined.Radiography is performed by different methods, including conventional X-ray radiography (RX) on film, computed radiography (CR), and direct digital imaging (DDI). Mammography is a special radiography technique for examining the breast, which has evolved rapidly in recent years with the following acquisition modifications: screen-film acquisition and reading in a light box, film digitization, 6 digital CR imaging, and more recently, full-field digital mammography (FFDM). Digital visualization has evolved with specialized 5-megapixel (MP) cathode-ray tube (CRT) monitors 7, 8 and specialized 5-MP liquid crystal display (LCD) monitors 9, 10 A recent study found no statistically significant differences in diagnostic accuracy for detecting nodules and calcifications between a 5-MP and a 3-MP specialized LCD medicalgrade monitor. 11 Emerging technologies are displacing existing ones, and the standards are changing. Currently, CR and FFDM with readings on specialized 5-MP LCD monitors is the standard clinical practice in mammography. LED monitors are replacing LCD monitors, and it is likely that the standard will further change. ...

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

“…These findings indicate that the digital devices provided comparable diagnostic accuracy to the reference treatment (FILM). In addition, these results are in agreement with our previous evaluation of displays (also including a medical-grade grayscale display) using images obtained with CR [17], as well as the evaluation of capture devices (including standard CR) visualized using a medical-grade grayscale display. Previous studies have compared film-screen mammography to digitized films [9][10][11], and similar to the present study, these reports found no significant differences in diagnostic accuracy.…”

“…The study of Krupinski [30] found also comparable result between medical-grade displays and consumer-grade displays, with more appreciated differences in the evaluation of subtle lesions, especially when the monitors have been used more than a year. In addition, in part of the our overall study when images from CR were included and all the images were reading using a medical-grade display, in addition to the consumer-grade displays of this evaluation, no differences were observed in detection of nodules and calcifications in screening mammograms when GSDF calibration or factory calibration was used [17].…”

“…The capture devices and displays evaluated in this study were also evaluated in our previous studies for chest radiography, and no significant differences in performance were noted in terms of detection between the capture devices [18][19][20] and displays evaluated [21][22][23]. In the current study, we evaluated whether the same devices could be used for mammography using low-cost solutions at both ends of the teleradiology chain, including capture with a specialized digitizer or digital camera and display with consumer-grade color LCD or LED displays.…”

“…These two displays were used with both the factory calibration and with the "Part 14: Grayscale Standard Display Function" (GSDF) DICOM standard calibration, as presented in our previous study [17].…”

“…When a medical-grade device was used at one end of the configuration (i.e., capture or display) in combination with a low-cost alternative, no significant differences were observed in comparison to film-screen readings [17].…”