An Enhanced Machine Learning Approach for Brain MRI Classification.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Biomedical Engineering

Background:

• Magnetic Resonance Imaging (MRI) is crucial for diagnosing disorders, but existing systems struggle with large datasets.
• Developing robust and efficient classification systems for comprehensive MRI datasets is essential.

Purpose of the Study:

• To develop a fast, reliable, and high-performing classification system for brain diseases using MRI images.
• To address the performance degradation of current systems when applied to large-scale MRI data.

Main Methods:

• Utilized global histogram equalization for image enhancement and noise reduction.
• Employed a symlet wavelet transform for optimal feature extraction from grayscale MRI images.
• Applied Linear Discriminant Analysis (LDA) for dimensionality reduction and logistic regression for classification.

Main Results:

• The proposed method achieved a classification accuracy of 96.6% on a diverse dataset.
• Demonstrated superior performance compared to existing state-of-the-art systems on large MRI datasets.
• Successfully classified 24 different brain disorders, including normal cases.

Conclusions:

• The developed system offers a robust and accurate approach for brain disease classification from MRI.
• The combination of symlet wavelet transform and LDA provides effective feature extraction and dimensionality reduction.
• This technique holds significant potential for improving diagnostic capabilities in medical imaging.