Statistical Wavelet based Adaptive Noise Filtering Technique for MRI Modality

Volume 4 Issue 4 December - February 2018

Research Paper

Statistical Wavelet based Adaptive Noise Filtering Technique for MRI Modality

C. Anjanappa*, H. S. Sheshadri**

* Assistant Professor, Department of Electronics and Communication Engineering, The National Institute of Engineering, Mysore, Karnataka, India.

** Professor, Department of Electronics and Communication Engineering, People's Education Society College of Engineering, Mandya, Karnataka, India.

Anjanappa, C., and Sheshadri, H. S. (2018). Statistical Wavelet based Adaptive Noise Filtering Technique for MRI Modality. i-manager’s Journal on Pattern Recognition, 4(4), 21-31. https://doi.org/10.26634/jpr.4.4.14130

Abstract

In this research work, wavelet domain method is designed to filter noise in medical images. This method adapts to various types of noise, which is dependent on the user or medical expert. Here, a single parameter can be used to balance the preservation of relevant details and the level of noise reduction. This method needs the subsequent information of the related image details across the resolution scales to perform a preliminary coefficient classification. The statistical distributions of the coefficients can be estimated by using preliminary coefficient classification that characterize the valuable image features and noise levels. Wavelet domain indicator is used to achieve the conversion to the image features and noise level. The experimental results demonstrated noise suppression in Magnet Resonance (MR) and Ultra Sound (US) images and its performance is validated by using quantitative and qualitative methods.