Three-dimensional anatomical model-based segmentation of MR brain images through Principal Axes Registration

Area of Science:

• Neuroimaging
• Medical Image Analysis
• Computational Anatomy

Background:

• Conventional anatomical atlases lack subject variability, limiting their effectiveness in segmenting and analyzing high-resolution MR brain images.
• Existing methods struggle to account for individual differences in brain structure, impacting the accuracy of neuroimaging analysis.

Purpose of the Study:

• To develop a method for creating 3D computerized composite brain models for a more accurate anatomical atlas.
• To improve model-based segmentation and analysis of brain structures by incorporating subject variability.
• To demonstrate the clinical utility of the composite model for assessing brain atrophy.

Main Methods:

• Development of 3D computerized composite models using registered MR brain images from multiple human subjects.
• Principal Axes Transformation for registering subject images.
• Model-based segmentation using the composite ventricular model for ventricles, sulci, and white matter lesions.

Main Results:

• A composite model of brain ventricles was developed from 22 human subjects.
• The composite model facilitated segmentation of ventricles, sulci, and white matter lesions in MR brain images.
• Automatic volumetric measurements revealed regional brain atrophy in chronic alcoholics compared to normal subjects.

Conclusions:

• Composite computerized anatomical atlases effectively represent brain structure variability.
• This approach enhances the accuracy of model-based segmentation and analysis of MR brain images.
• The method shows clinical usefulness in quantifying brain atrophy, particularly in conditions like chronic alcoholism.