Segmentation
and Flattening Software
|
|
|
|
Wandell Lab (Stanford) Wade and Taylor
Productions |
Download
Follow these links to find documentation on how to use mrGray and how to use mrFlatMesh.
Description
The segmentation and flattening tools distributed at this site permit you to identify the gray matter in visual cortex and then to produce a visualization of the gray matter in a single plane. We developed these tools in order to help people see the spatial relationship between signals in cortex. We found understanding the spatial relationship complex when using multi-slice or glass-brain formats.
Papers describing the ideas behind the Windows software distributed here can be found at in the Visualization section of my recent functional mri publications. If you use our software, we would appreciate a citation of
- Visualization
and Measurement of the Cortical Surface.
B. Wandell, S. Chial and B. Backus (2000). Journal of Cognitive Neuroscience, vol. 12, no. 5. pp. 739-52.
This paper describes the methods used in the unfolding software that we distribute (see below). It also provides an overview of the principles, problems, and related methods that can be found at other sites. - Creating
Connected Representations of Cortical Gray Matter for Functional MRI
Visualization
Teo, Sapiro and Wandell (1997). IEEE Med. Transactions, vol. 16 (6), 852-863.
This paper describes the algorithm for obtaining a segmented and connected gray matter representation from T1 weighted MR images. The algorithm is implemented in the software package we distribute (see below).
In addition to the tools provided here, you may be interested in distributions from several other groups. If you work in a Windows or Linux environment, you might be interested in the commercial package from Brain Innovation (Rainer Goebel). There are two distributions specifically for the unix platform: FreeSurfer from the Anders Dale, Bruce Fischl, and the MGH group; or the Caret, Flatmorph and Xanat tools from David Van Essen’s team at Washington University, St. Louis. Those sites contain papers describing their objectives and methods.
There are two steps involved in creating flattened cortical representations. First, one must identify the surface that will be flattened. This process is called segmentation. One begins with anatomical images that have good contrast between white matter and gray matter. The segmentation tool, in this case mrGray, identifies the white and gray matter and from this information finds a surface either at the boundary between white and gray or within the gray matter.
Second, one must use an algorithm to achieve the computational flattening. This unfolding tool, in this case mrFlatMesh, defines a method for placing the points along the cortical surface (which meanders through 3-space) at locations within the plane. The method implemented in mrFlatMesh (by Alex Wade) uses a very efficient algorithm to produce a flattened image guaranteed to have no twists in the surface.
We have also identified software tools (in Matlab) for automatically segmenting visual areas from phase-encoded retinotopic mapping experiments. This code is now part of the mrLoadRet/mrVista distribution. The principles are described in
- Visual field representations and
locations of visual areas V1/2/3 in human visual cortex
R. F. Dougherty, V. M. Koch, A. A. Brewer, B. Fischer, J. Modersitzki and B. A. Wandell (2003).
Visual field representations and locations of visual areas V1/2/3 in human visual cortex.
Journal of Vision , 3(10),586-598, DOI 10.1167/3.10.1.
Good luck with your work.
Brian Wandell
Distribution history:
§
April, 1996:
mrGray-1.0 (Teo) and mrUnfold-2.0 (Wandell, Engel)
§
April 29, 1999:
mrUnfold-4.0 distribution (Wandell, Chial , Georghiades,) and Windows form
of mrGray (R.Taylor)
§
June 14, 2000:
mrGray-2.0 (R. Taylor) distribution that includes 3D rendering
§
May, 2001:
mrUnfold-5.0 (Alex Wade) rapid flattening and mrGray-3.0 (Wade modifications)