New Transfer Theory Relationships for Signal and Noise Analyses of X-ray Detectors

X-ray mammography is currently the most reliable method available for the detection of breast cancer in screening programs, but it still does not detect all cancers. A great deal of research effort over the past several decades has been directed towards the development of better and more effective imaging systems. These new systems must be designed carefully to ensure they can produce images of the highest quality possible. Fourier-based linear-systems transfer theory is often used to develop theoretical models of the signal and noise performance of new system designs. While it has been used successfully in a number of new system designs, only relatively simple systems can be analyzed using this approach. We are developing new Fourier-based transfer relationships that will extend the capabilities of linear-systems theory so that it can be used in the design of increasingly complex systems. The most important outcome of the first year of progress has been development of the idea of parallel cascaded of amplified point processes. Using it, linear-systems transfer theory can be used to predict the detective quantum efficiency DQE during the design of complex x-ray detectors being developed for digital mammography, to ensure optimal design of these detectors that will maximize image quality for any specified radiation dose to the patient.