Introduction and Acknowledgements

This work was partially funded by CAPES and CNPq, a Brazillian Agencies. Information on CAPES can be obtained on the CAPES website and CNPq website.
Author: Antonio Carlos da S. Senra Filho, CSIM Laboratory (University of Sao Paulo, Department of Computing and Mathematics)
Contact: Antonio Carlos da S. Senra Filho <email>acsenrafilho@usp.br</email>

CSIM Laboratory

University of Sao Paulo

CNPq Brazil

CAPES Brazil

Extension Description

This extension provides a image segmentation and local image contrast enhancement approaches in order to detect abnormal white matter voxels signals in magnetic resonance images. At moment, there are available the LS Segmenter (specific for hyperintense Multiple Sclerosis lesion segmentation on T2-FLAIR images) and LS Contrast Enhancement (specific to increase the contrast of abnormal voxels of T2-FLAIR images) modules. The LS Contrast Enhancement module can be used as an image contrast enhancement pre-processing step before a future lesion segmentation algorithm (which could be used any other approach, such as LesionTOADS, OASIS or LST). The LS Segmenter module implements a hyperintense T2-FLAIR lesion segmentation based on the algorithm published in the paper^[1].

NOTE: The Logistic Contrast Enhancement, Weighted Enhancement Image Filter and Automatic FLAIR Threshold modules are only supporting CLI methods added in the extension to calculate specific parts of the segmentation procedure. For this reason, these modules are not supposed to be used alone. Please, use the LS Contrast Enhancer or LS Segmenter modules only.

Modules

Use Cases

Most frequently used for these scenarios:

Use Case 1: Hyperintense Multiple Sclerosis (MS) lesions segmentation
- T2-FLAIR images are usually applied to MS diagnosis in order to detect hyperintense MS lesions. In this case, the LS Segmenter module can be useful.
Use Case 2: Increase contrast in abnormal voxels in white matter tissue
- There are some lesion segmentation approaches that relies on the voxel intensity level presented in the white matter tissue, where the LS Contrast Enhancer module can be helpful to increase the contrast between lesions and surrounding brain tissues (mainly normal appearing white matter - NAWM).

T2-FLAIR image from a MS patient
T2-FLAIR hyperintense lesion segmentation provided by the LS Segmenter module
Lesions and White Matter 3D representation
T2-FLAIR image with original lesion contrast
T2-FLAIR image with hyperintense MS lesion contrast enhanced by the LS Contrast Enhancer module