Stationary Wavelet Transform and AdaBoost with SVM Based Pathological Brain Detection in MRI Scanning.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Biomedical Engineering

Background:

• Accurate classification of pathological brain conditions from normal Magnetic Resonance Imaging (MRI) is crucial for diagnosis.
• Existing methods for brain image analysis often face challenges in feature extraction and classification accuracy.

Purpose of the Study:

• To develop and validate an automated system for classifying pathological brain images from normal ones using MRI scans.
• To enhance the performance of brain image classification through advanced image processing and machine learning techniques.

Main Methods:

• Utilized contrast limited adaptive histogram equalization for enhancing diseased regions in brain MR images.
• Employed two-dimensional stationary wavelet transform (SWT) for feature extraction, focusing on energy and entropy from level-2 SWT coefficients.
• Implemented a feature selection process using a symmetric uncertainty ranking filter to identify relevant and uncorrelated features.
• Developed a classification model combining AdaBoost with Support Vector Machine (SVM) as the base classifier.

Main Results:

• The proposed system demonstrated superior performance compared to existing schemes in terms of classification accuracy and the number of features used.
• Achieved ideal classification results on Dataset-66 and Dataset-160.
• Attained a high accuracy of 99.45% on Dataset-255.
• Validated through 5 runs of k-fold stratified cross-validation.

Conclusions:

• The developed automated system effectively segregates pathological from normal brain MR images.
• The combination of image enhancement, wavelet transform features, and AdaBoost-SVM classifier offers a robust and accurate approach for brain MRI analysis.
• The system's performance indicates its potential for clinical application in diagnosing brain pathologies.