MRI Brain Images Classification: A Multi-Level Threshold Based Region Optimization Technique

Area of Science:

• Medical image processing
• Computer-aided diagnosis
• Neuroimaging analysis

Background:

• Medical image processing, particularly Magnetic Resonance Image (MRI) analysis, is crucial for diagnosing brain conditions.
• Manual segmentation of infected regions in MR brain images is time-consuming, subjective, and heavily reliant on expert experience.
• Existing methods face challenges in efficiency and accuracy for complex medical image segmentation.

Purpose of the Study:

• To develop an efficient computer-aided technique for brain tumor segmentation.
• To enhance the accuracy and reduce the complexity of classifying normal versus abnormal tissues in MR brain images.
• To introduce the Threshold Based Region Optimization (TBRO) method for improved medical image analysis.

Main Methods:

• Proposed a novel Threshold Based Region Optimization (TBRO) algorithm for brain tumor segmentation.
• Utilized Magnetic Resonance (MR) brain images as the data source for the classification system.
• Implemented the technique using MATLAB software for detection, extraction, and classification of tumors.

Main Results:

• The TBRO technique achieved high classification performance metrics.
• Experimental results demonstrated 96.57% accuracy in classifying tissues.
• Achieved 94.6% specificity and 97.76% sensitivity, indicating robust performance.

Conclusions:

• The proposed Threshold Based Region Optimization (TBRO) method effectively segments brain tumors from MR images.
• The computer-aided approach significantly improves classification accuracy and reduces reliance on manual expert analysis.
• TBRO offers a promising solution for efficient and accurate brain tumor detection in medical imaging.