Identification and classification of brain tumor MRI images with feature extraction using DWT and probabilistic...

Area of Science:

• Medical Imaging
• Computer Vision
• Biomedical Engineering

Background:

• Accurate brain tumor identification in MRI is challenging due to complex anatomy and image noise.
• Traditional methods for segmentation and detection are often time-consuming and labor-intensive.

Purpose of the Study:

• To develop an efficient and accurate automated technique for brain tumor segmentation and detection in MRI images.
• To reduce the complexity and improve the performance of tumor identification processes.

Main Methods:

• Noise removal and Gray-Level Co-occurrence Matrix (GLCM) feature extraction.
• Discrete Wavelet Transform (DWT)-based region growing segmentation.
• Morphological filtering for post-segmentation noise reduction.
• Probabilistic Neural Network (PNN) classifier for tumor detection.

Main Results:

• The proposed technique achieved nearly 100% accuracy in distinguishing between normal and abnormal brain tissues.
• Demonstrated high effectiveness in identifying tumor locations within MRI scans.
• Significantly reduced the complexity associated with manual tumor detection.

Conclusions:

• The integrated approach of DWT-based segmentation, morphological filtering, and PNN classification offers a highly accurate solution for brain tumor detection in MRI.
• This automated method enhances diagnostic efficiency and reliability in neuro-oncology.