Transform-Based Denoising and Enhancement in Medical X-Ray Imaging

Physical and clinical constraints of the imaging process often degrade the quality of medical images. A typical case is low-dose X-ray image acquisition, where the signal-to-noise ratio is limited due to X-ray quantum noise (photon-limited imaging). This paper discusses and quantitatively evaluates spectral transform-based methods for denoising and enhancement of such images. We examine algorithms based on the DFT, the Modulated Lapped Transform (MLT) and a new lapped transform termed the Lapped Directional Transform (LDT).