Denoising Mammographic Images Using ICA

Abstract

Abstract. Digital mammographic image processing often requires a previous application of filters to reduce the noise level of the image while preserving important details. This may improve the quality of digital mammographic images and contribute to an accurate diagnosis. Denoising methods based on linear filters cannot preserve image structures such as edges in the same way that methods based on nonlinear filters can do it. Recently, a nonlinear denoising method based on ICA has been introduced [1,2] for natural and artificial images. The functioning of the ICA denoising method depends on the statistics of the images. In this paper, we show that mammograms have statistics appropriate for ICA denoising and we demonstrate experimentally that ICA denoising is a suitable method to remove the noise of digitised mammographys.

Comparing methods to denoise mammographic images

“…The choice of a shrinkage function depends on the statistical distribution of each sparse component [5]. It has been shown [10] that the statistical distributions of the independent components of mammographic images are appropriated to apply the shrinkage algorithm introduced in [4,5].…”

“…It is possible applying the ICA denoising method [4,5] to mammographic images [10]. The density functions analysis of the independent components of mammographic images show that these distributions are suitable to apply the ICA denoising method.…”