Joint Anisotropic LMMSE Filter for Stationary Rician noise removal in DWI

This module reduces Rician noise on DWI. It filters the image in the mean squared error sense using a Rician noise model. The N closest gradient directions to the direction being processed are filtered together to improve the results: the noise-free signal is seen as an n-dimensional vector which has to be estimated with the LMMSE method from a set of corrupted measurements. The covariance matrix of the noise-free vector and the cross covariance between this signal and the noise have to be estimated, which is done taking into account the image formation process.

All these estimations are performed as sample estimates in a 'shaped neighborhood' defined by the weights extracted from the structural similarity of the voxels following the same idea as in the Non-Local Means filter.

This C++ code can be compiled as either a standalone (ITK required. Versions 3 and 4 are supported) or as a CLI module for 3-D Slicer (versions 3 and 4). Only "nhdr/nrrd" DWIs are supported.

Execution Options

Download Now:


3D Slicer License
Intended Audience:
Natural Language:
Operating System:
Programming Language:
Supported Data Format:


build requires:
is a plugin for: