OBJECTIVE ASSESSMENT OF PHANTOM IMAGE QUALITY IN MAMMOGRAPHY - A FEASIBILITY STUDY

The need for test objects in mammography quality control programmes to provide an objective measure of image quality pertinent to clinical p roblems is well documented. However, interobserver variations may be g reater than the fluctuations in image quality that the quality control programme is seeking to detect. We have developed a computer algorith m to score a number of features in the Leeds TOR(MAX) mammography phan tom. Threshold scoring techniques have been applied in the first insta nce; scoring schemes which utilize measures such as signal-to-noise ra tio and modulation have also been formulated. This fully automatic alg orithm has been applied to a set of 10 films which have been digitized at 25 micron resolution using a Joyce-Loebl scanning microdensitomete r. The films were chosen retrospectively from quality control test fil ms to demonstrate: (a) a range of optimized imaging systems, and (b) v ariation from the optimum. The performance of the algorithm has been c ompared with that of five experienced observers, and has been shown to be as consistent as individual observers, but more consistent than a pool of observers. Problems have been encountered with the detection o f small details, indicating that a more sophisticated localization tec hnique is desirable. The computer performs more successfully with the scoring scheme which utilizes the full imaging information available, rather than with the threshold-determined one. However, both the obser vers and the computer algorithm failed to identify the non-optimum fil ms, suggesting that the sensitivity of the TOR(MAX) test object may no t be adequate for modern mammography imaging systems.