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

Dinh Hoan Trinh1, Marie Luong2, Jean-Marie Rocchisani3, Canh Duong Pham4, Françoise Dibos1, Linh-Trung Nguyen5
1LAGA, Villetaneuse, France
2L2TI, Villetaneuse, France
3Hopital Avicenne-Medecine Nucleaire, Bobigny, France
4CIID, Vietnam Academy of Science and Technology, Hanoi, Vietnam
5Fac. Electronics & Telecom., Univ. Eng. & Tech., VNU-Hanoi, Vietnam

Tóm tắt

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.

Từ khóa

#Noise #Training #Noise reduction #Biomedical imaging #Noise measurement #Image denoising #Rician channels

Tài liệu tham khảo

10.1109/ISBI.2009.5193135

saunders, 1998, Ridge regression learning algorithm in dual variables, Machine Learning Proc 15th Int Conf (ICML 98), 515

10.1023/A:1009715923555

10.1088/0031-9155/55/16/N02

feng, 2002, Multiscale principal components analysis for image local orientation estimation, Conference Record 36th Asilomar Conf Signals Systems and Computers (), 1, 478

10.1109/TIP.2003.819861

10.1080/00401706.1979.10489751

wiest-daesslé, 2008, Rician noise removal by non-local means filtering for low signal-to-noise ratio MRI: Applications to DT-MRI, Proc 11th Int Conf Medical Image Computing and Computer-Assisted Intervention (MICCAI 2008), 171, 10.1007/978-3-540-85990-1_21

10.1016/j.media.2008.02.004

10.1016/0167-2789(92)90242-F

thaipanich, 2010, An adaptive nonlocal means scheme for medical image denoising, Proc SPIE Medical Imaging 2010 Image Processing, 7623, 76 230m, 10.1117/12.844064

10.1109/TMI.2003.809588

gilboa, 2003, Texture preserving variational denoising using an adaptive fidelity term, Proc of IEEE Workshop on Variational Geometric and Level Set Methods in Computer Vision, 137

10.1137/040616024

10.1002/mrm.1910340618

10.1016/j.imavis.2008.06.006