Related Experiment Video
Updated: Jul 31, 2025
06:48
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
Published on: January 7, 2019
9.0K
Brain image fusion-based tumour detection using grey level co-occurrence matrix Tamura feature extraction with
1Department of ECE, Sathyabama Institute of Science and Technology, Chennai 600119, India.
Mathematical Biosciences and Engineering : MBE
|May 10, 2023
Summary
This study introduces an efficient brain tumor detection method using fused MRI and CT images. The approach accurately classifies tumors as benign or malignant, aiding in patient health monitoring.
Area of Science:
- Medical Image Analysis
- Artificial Intelligence in Healthcare
- Biomedical Engineering
Background:
- Brain tumor detection is a critical yet challenging task in medical image analysis.
- Existing methods often require sophisticated techniques for accurate diagnosis.
- Integrating multiple imaging modalities can enhance detection efficiency and accuracy.
Purpose of the Study:
- To develop an automated system for accurate brain tumor detection and segmentation.
- To improve the efficiency of medical image analysis through image fusion and feature extraction.
- To classify detected tumors as benign or malignant using a supervised learning approach.
Main Methods:
- Preprocessing and fusion of Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) images using Stationary Wavelet Transform (SWT) and Inverse Stationary Wavelet Transform (ISWT).
- Feature extraction from fused images using the Gray-Level Co-occurrence Matrix (GLCM)-Tamura method.
- Classification of tumors using a Backpropagation Network (BPN) classifier trained on segmented tumor regions (benign/malignant) via k-means clustering.
- Implementation of a software system for patient health status notification via Global System for Mobile Communications (GSM).
Main Results:
- The integrated approach demonstrated effective tumor segmentation and classification.
- Experimental analysis yielded quantifiable metrics including accuracy, precision, recall, F1-score, Root Mean Square Error (RMSE), and Mean Average Precision (MAP).
- The system successfully distinguished between benign and malignant tumor regions.
Conclusions:
- The proposed method offers an efficient and accurate solution for brain tumor detection and classification.
- Image fusion and advanced feature extraction techniques significantly enhance diagnostic capabilities.
- The developed system facilitates timely patient notification, potentially improving healthcare outcomes.
