Denoising of mammographic images from quantum noise in wavelet domain

Abstract

Telemedicine is the new trends in Health care services. Advances in Electronics, Imaging Techniques and Communication Engineering help a lot in early diagnosis and fast recovery of some fatal diseases. As an example case of breast cancer is considered here. Digital Mammographic images are used for diagnosis of Breast cancer. But they are contaminated by quantum noise during acquisition due to the nature of low energy photons used for Mammographic imaging modality. To remove the Quantum Noise in mammographic images, use of Discrete Wavelet Transforms (DWT) is gaining momentum due to unique properties of sparcity and multiresolution and easy implementation of DWTs with different Digital Filters. Thresholding techniques such as VisuShrink, BayesShrink, NeighShrink and Modified Neighbourhood are used in this paper and three different wavelets as Haar, Db4 and Sym4 has been used. The Performance Metrics such as Peak Signal to Noise Ratio(PSNR), Mean square Error(MSE), Structural Similarity Index (SSIM) and Edge Preserving Index (EPI) are used to evaluate the performance of Denoising algorithms.