Abstract

As the medical X-ray image shifted from the analog system to the digital system, not only the format of the image changed but also the form of diagnosis changed. We will examine this problem with mammography as an example. Mammography has characteristics of high sharpness and high contrast to detect the microcalcifications and masses. In the era when using the intensifying screen-film systems, sharpness could be secured but the contrast rise was limited. Low energy X rays were used to obtain high contrast. However, this choice has led to excessive radiation dose to the breast. Even when the mammography system shifted to the digital system, the energy of the X rays used remained low. Was this choice correct? The light and shade of the analog system could be evaluated by contrast value, but in the digital system the contrast is variable by image processing and cannot be used an index of evaluation. Therefore, the signal-to-noise ratio （SNR）or contrast-to-noise ratio（CNR）has been proposed instead of the index of contrast for digital systems. A central issue in the study of radiological image quality is to obtain knowledge about the influence of physical image quality on diagnosis. We introduce a study of the correlation between several CNR models and the actual signal detection rate. We are also studying for the purpose of developing new digital mammography that uses a Cadmium Telluride（CdTe）series photon counting detector. This system uses higher energy than the typical mammographic energy, using a tungsten target. We discussed the possibility of dose reduction and new diagnosis by substance identification when using our proposed system. Through development of this systems, we consider the possible appearance of mammography in the new digital age.