A novel brain MRI classification framework integrating tuned single scale retinex and empirical wavelet entropy...

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Biomedical Engineering

Background:

• Brain tumors are a leading cause of death in adults worldwide.
• Early diagnosis of brain tumors is crucial for improving patient prognosis.
• Magnetic Resonance Imaging (MRI) is a key non-invasive diagnostic tool for visualizing brain tissues.

Purpose of the Study:

• To propose a novel framework for brain MRI image classification.
• To enhance image quality and extract informative features for accurate tumor detection.
• To evaluate the performance of the proposed classification framework.

Main Methods:

• Image enhancement using Tuned Single-Scale Retinex (TSSR).
• Adaptive decomposition of enhanced images using Empirical Wavelet Transform (EWT).
• Extraction of energy and entropy features (Shannon, Tsallis) and classification using Support Vector Machine (SVM) and LPBoost.

Main Results:

• Achieved a classification accuracy of 96.43% on a binary-class brain MRI dataset.
• Demonstrated a True Positive Rate (TPR) of 100% and a True Negative Rate (TNR) of 77.78%.
• Outperformed several existing state-of-the-art methods in brain MRI classification.

Conclusions:

• The proposed framework offers a robust and generalizable solution for medical image analysis.
• EWT-based statistical features and TSSR-enhanced image quality are key contributions.
• The method shows significant potential for improving early brain tumor diagnosis through MRI analysis.