A fusion method of Gabor wavelet transform and unsupervised clustering algorithms for tissue edge detection

Area of Science:

• Medical Imaging
• Image Processing
• Computer Vision

Background:

• Accurate edge detection is crucial for medical image analysis, aiding in diagnosis and treatment planning.
• Traditional methods often struggle with noise and low contrast inherent in medical imaging modalities like CT and MRI.
• Unsupervised clustering offers a data-driven approach to image segmentation, but requires effective feature extraction.

Purpose of the Study:

• To develop and evaluate two novel unsupervised edge detection methods for medical images.
• To leverage the noise suppression and edge enhancement capabilities of Gabor Wavelet Transform (GWT).
• To integrate GWT with k-means and Fuzzy c-means (FCM) clustering for robust image segmentation.

Main Methods:

• Gabor Wavelet Transform (GWT) was applied to enhance image features and reduce noise.
• K-means and Fuzzy c-means (FCM) clustering algorithms were employed for image segmentation.
• The proposed methods were validated on Computed Tomography (CT), Magnetic Resonance Imaging (MRI), and phantom images.

Main Results:

• The integrated GWT and clustering methods demonstrated successful edge detection in medical images.
• Effective noise suppression was achieved, preserving critical edge information.
• The algorithms proved robust even in the presence of significant image noise.

Conclusions:

• The proposed GWT-integrated clustering methods offer a significant advancement in medical image edge detection.
• These techniques enhance the reliability of image analysis, particularly for noisy CT and MRI scans.
• The study validates the effectiveness of combining wavelet transforms with unsupervised clustering for medical image processing.