Noise Reduction in Phase-Contrast Mammography Images with Synchrotron Radiation

Abstract

Phase-contrast mammography using synchrotron beams (PCI_SR) has shown quantitatively better results in terms of image contrast and detection of small details with respect to conventional digital techniques. However, this new technique has shown higher levels of random noise than common digital techniques. The objective of this work was to reduce noise in PCI_SR images to levels, similar to those of hospital equipment without significantly affecting the spatial resolution, contrast and visibility of fine details. Five filters: Wiener, Gauss, bilateral and coif and biord wavelets families, were used independently to reduce image noise. Image quality before and after filtering was estimated using objective indexes. The results showed increases in Signal-to-Noise Ratio (SNR) without impairing other quality indexes; which demonstrates the superiority in image quality of this innovative technique for the detection of small lesions such as microcalcifications and low contrast masses.