Brain MR image segmentation using NAMS in pseudo-color.

Area of Science:

• Biomedical Imaging
• Computer Vision
• Medical Image Analysis

Background:

• Image segmentation is vital for biomedical applications, particularly for analyzing and planning surgeries using brain Magnetic Resonance (MR) images.
• Accurate segmentation of brain MR images aids in representing the image with homogeneous regions, moving beyond pixel-level analysis.

Purpose of the Study:

• To propose a novel pseudo-color based segmentation method for MR brain images.
• To enhance the precision and visual distinction of brain MR image segmentation.
• To reduce data redundancy and storage requirements for segmented images.

Main Methods:

• Introduction of the Non-symmetry and Anti-packing Model with Squares (NAMS) for image representation using sub-patterns.
• Conversion of grayscale brain MR images into pseudo-colored images to improve tissue contrast.
• Segmentation of the pseudo-colored MR images using the NAMS model.

Main Results:

• The proposed pseudo-color segmentation method using NAMS demonstrates superior performance compared to existing brain MR image segmentation techniques.
• The method achieves precise segmentation of brain MR images.
• Significant reduction in data storage is achieved through the NAMS model's data redundancy reduction capabilities.

Conclusions:

• The NAMS based pseudo-color segmentation method offers an excellent approach for segmenting brain MR images.
• This technique improves segmentation accuracy and provides better visual perception.
• The method is efficient in terms of both precision and storage savings for biomedical applications.