Jump to: navigation, search

In this section . . .


Designed specifically for medical researchers, MIPAV concentrates on providing those researchers with the tools needed to do their work. It reads image files of many different formats and allows images to be displayed and measured using the most meaningful method to achieve research goals. MIPAV's flexibility becomes apparent when its capabilities are expanded and fine tuned through the development of plug-in programs that tailor solutions to meet specific requirements.

Using MIPAV to display, label, and measure brain components in Talairach space demonstrates both: MIPAV's native ability to display and measure brain images in Talairach space and the tailoring provided through the Talairach Transformation wizard and the FANTASM (Fuzzy and Noise Tolerant Adaptive Segmentation Method) plug-in programs, developed by the Johns Hopkins University.


In 1988 Jean Talairach and Pierre Tournoux developed a three-dimensional proportional grid system that can be used to identify and measure brains from any number of patients despite the variability of brain sizes and proportions. The premise of the system is that brain components that cannot be seen or identified can be defined in relation to other anatomic cerebral structures. In the Talairach system, the anterior commissure (AC) and posterior commissure (PC) are the structures from which the system of reference is developed.

The Talairach system establishes the maximal dimensions of the brain in three planes of space: x, y and z:

AC-PC line (X axis)-A horizontal line running through the anterior and posterior commissures.
VCA line (verticofrontal line, or Y axis)-A vertical line passing through the anterior commissure
Midline (Z axis)-A line forming the interhemispheric sagittal plane

Often referred to as the "origin," the anterior commissure is commonly used to describe structures. For example, a structure is described as "AC 13 mm" for the frontal lobe or "AC - 35 mm" for the occipital pole. These descriptions assume that the anterior commissure is in the positive direction. However, the Talairach system does not use positive and negative directions. Instead, it labels quadrants according to number and letters (Figure 1). The AC-PC line defines the horizontal plane, the VCA line defines the vertical plane, and the midline defines the depth plane. Because the anterior commissure and posterior commissure do not occur in the same axial slice, reslicing is necessary to put the brain into Talairach space.

This technical guide explains how to install and use two MIPAV plug-in programs-the Talairach Transformation wizard and FANTASM to:

Create the x, y, and z planes of space in an image of a brain
Transfer Talairach labels to an image of a brain
Measure brain components in Talairach space

Figure 1. Talairach space: (A) Quadrants labeled by number and letters and (B) horizontal, vertical, and depth planes


Talairach Transformation wizard

The Talairach Transformation wizard is a plug-in program for MIPAV that performs a semimanual transformation of image datasets of the brain to Talairach (stereotaxic) coordinates, providing atlas-based labeling. The Talairach coordinates allow researchers to easily identify subregions of the brain and measure their volume. It includes labels for 148 different substructures of the brain at various scales, obtained from the [ ]Talairach Daemon database, along with a set of volumetric images of the labels.


The FANTASM plug-in program is a different version of the Fuzzy C-mean algorithm for segmenting 2D and 3D images. It incorporates a spatial constraint that requires neighboring pixels to be similar and reduces the noise effect obtained with the Fuzzy C-mean algorithm. It can deal with outliers. Plans for a future version of FANTASM incorporates inhomogeneity correction.


ICBM atlas created by the International Consortium on Brain Mapping (ICBM), automatic

Jean Talairach and Pierre Tournoux, Co-Planar Stereotaxic Atlas of the Human Brain, Thieme Medical Publishers, New York, 1988.
Neva Chernizasky, Medical Imaging: Orientation, Paper prepared for Matthew McAuliiffe, Ph.D. Center for Information Technology, National Institutes of Health, August 31, 2001.
Dzung L. Pham, "Spatial Models for Fuzzy Clustering," Computer Vision and Image Understanding, vol. 84, pp. 285-297, 2001.
Pierre-Louis Bazin, Dzung L. Pham, William Gandler, and Matthew McAuliffe. "Free Software Tools for Atlas-based Volumetric Neuroimage Analysis," to be published in the Proceedings of the SPIE Medical Image 2005 Conference, The International Society for Optical Engineering (SPIE), Bellingham, Washington, 2005.