Denoising and Enhancement of Mammographic Images Under the Assumption of Heteroscedastic Additive Noise by an Optimal Subband Thresholding

Abstract

Mammographic images suffer from low contrast and signal dependent noise, and a very small size of tumoral signs is not easily detected, especially for an early diagnosis of breast cancer. In this context, many methods proposed in literature fail for lack of generality. In particular, too weak assumptions on the noise model, e.g., stationary normal additive noise, and an inaccurate choice of the wavelet family that is applied, can lead to an information loss, noise emphasizing, unacceptable enhancement results, or in turn an unwanted distortion of the original image aspect. In this paper, we consider an optimal wavelet thresholding, in the context of Discrete Dyadic Wavelet Transforms, by directly relating all the parameters involved in both denoising and contrast enhancement to signal dependent noise variance (estimated by a robust algorithm) and to the size of cancer signs. Moreover, by performing a reconstruction from a zero-approximation in conjunction with a Gaussian smoothing filter, we are able to extract the background and the foreground of the image separately, as to compute suitable contrast improvement indexes. The whole procedure will be tested on high resolution X-ray mammographic images and compared with other techniques. Anyway, the visual assessment of the results by an expert radiologist will be also considered as a subjective evaluation.

Mammogram denoising by curvelet transform based on the information of neighbouring coefficients

“…The prime steps involved in that work are the mammogram enhancement by homomorphic filtering, detection of the region of interest by Local Seed Region Growing algorithm, feature extraction by the spherical wavelet transform and the classification of the breast masses by the classifier, Support Vector Machine. A suitable wavelet thresholding suggested by Mencattini et al [5] in connection to the discrete dyadic wavelet transform relates all the denoising and enhancement related parameters to the noise variance and cancer symptoms. The background and the foreground of the mammograms are properly segmented to facilitate the computation of contrast enhancement index.…”

Poisson Noise Removal from Mammogram Using Poisson Unbiased Risk Estimation Technique

“…One such work by Görgel et al [2] demonstrates the denoising of the mammograms by the wavelet transform and homomorphic filtering. Again, mammogram denoising and enhancement methodology proposed by Mencattini et al [3] is based on the optimal thresholding which relates all the denoising and contrast enhancement parameters to the noise variance and the extent of cancer signs. A recent investigation [4] developed on the spherical wavelet transform for feature extraction and Local Seed Region Growing algorithm for region of interest identification describes the detection and classification of suspect breast masses as benign and malignant.…”

Improving Image Quality in Medical Images Using a Combined Method of Undecimated Wavelet Transform and Wavelet Coefficient Mapping

“…In order to achieve shift invariance, researchers have proposed the undecimated DWT (UDWT) [ 8 – 10 ]. Mencattini et al reported a UDWT-based method for the reduction of noise in mammographic images [ 11 ]. The reported method was robust and effective.…”