Brain Tissue Segmentation from Magnetic Resonance Brain Images Using Histogram Based Swarm Optimization Techniques

Area of Science:

• Medical image analysis
• Computational neuroscience
• Artificial intelligence in healthcare

Background:

• Automated brain tissue segmentation is crucial for efficient diagnosis.
• Magnetic Resonance Imaging (MRI) is a key modality for brain imaging.
• Existing methods face challenges in time complexity and computational efficiency.

Purpose of the Study:

• To propose an automated brain tissue segmentation method.
• To reduce time complexity and improve computational efficiency in diagnostic processes.
• To enhance the accuracy of brain tissue segmentation in MRI.

Main Methods:

• The proposed method involves preprocessing and segmentation using Histogram based Swarm Optimization techniques.
• It was validated on T1-Weighted and T2-Weighted brain MRI datasets, including data from IBSR, MIRIAD, and SBC Scan Center.
• The technique was evaluated using Jaccard (JC) and Dice (DC) coefficients.

Main Results:

• The Histogram based Darwinian Particle Swarm Optimization (HDPSO) method demonstrated superior performance.
• HDPSO outperformed other swarm optimization techniques (HPSO, HFODPSO) and conventional methods (AMAP, BMAP, MAP, ML, TK-Means).
• Quantitative evaluation confirmed the effectiveness of the proposed segmentation approach.

Conclusions:

• The developed HDPSO method significantly improves automated brain tissue segmentation.
• This approach offers enhanced computational efficiency and accuracy for brain image analysis.
• The findings suggest HDPSO as a promising technique for clinical diagnostic applications.