MR image denoising using nonlinear regression and Fuzzy C-Means clustering

Magnetic Resonance (MR) imaging is useful for medical diagnosis. However, MR images are often corrupted by Rician noise, leading to undesirable visual quality. Based on the fact that many images can be acquired at nearly the same location, this paper proposes a novel learning method for the reduction of Rician noise using nonlinear ridge regression with a training set established from a set of given standard images. In addition, Fuzzy C-Means (FCM) is used for the classification of the training set. Experimental results show that our method outperforms some state-of-the-art methods.