Anatomically Consistent Enhancement (ACE)

This algorithm operates on tissue memberships derived from the TOADS brain tissue classification algorithm. It takes as input gray matter (GM) and white matter (WM) membership functions and "enhances" the gray matter membership in deep sulcal regions where voxels are likely to contain partial volume of gray matter and cerebral spinal fluid (CSF). This enhancement is done by fast-marching the WM membership with a variable speed to obtain a GM "skeleton." The skeleton is subtracted from the GM membership to emphasize sulcal regions (see reference [1] for details).

Input Types

You should be able to apply this algorithm 3D images .

ACE Input Parameters

Gray Matter

A float-valued volume containing gray matter membership values (scaled between 0 and 255).

White Matter

A float-valued volume containing white matter membership values (scaled between 0 and 255).

Intensity Threshold

Specifies convergence when all membership functions (of the fuzzy C cluster segmentation) over all pixel locations j change by less than this tolerance value between two iterations. The default value is 126.9.

ACE Outputs

Enhanced GM

The enhanced GM membership with "emphasized" sulcal regions.

Skeleton

The skeleton of the GM membership used to estimate the locations of CSF.

Thinned Skeleton

A thinned version of the skeleton. This thinned version is usually used since the raw skeleton is typically too thick.

Example Usage

Example input images.

 Input Gray matter membership

Gray matter membership

 Input White matter membership

White matter membership

Output images

 Input Image

Enhanced gray matter image.

 Input Image

Skeleton image.

 Input Image

Thinned skeleton image.

References


[1] X. Han, D. L. Pham, D. Tosun, M. E. Rettmann, C. Xu, and J. L. Prince, "CRUISE: cortical reconstruction using implicit surface evolution.," NeuroImage, vol. 23, no. 3, pp. 997-1012, 2004.