Noise removal methods for high resolution MRI

It is well-known that the noise in magnetic resonance images obeys a Rician distribution. Unlike additive Gaussian noise, Rician noise is signal-dependent and consequently separating signal from noise is a difficult task. Rician noise is especially problematic in high resolution low signal-to-noise ration regimes where it not only causes random fluctuations, but also introduces a signal-dependent bias to the data that reduces image contrast. This paper reports on a novel wavelet-domain filter that removes Rician noise contamination in MR images. The new method is especially well-suited for high-resolution, low SNR imaging.