Identification of masses in digital mammogram using gray level co-occurrence matrices

Area of Science:

• Medical Imaging
• Computer-Aided Diagnosis
• Biomedical Engineering

Background:

• Digital mammography is crucial for early breast cancer detection.
• Automated analysis of mammograms can improve diagnostic efficiency.
• Computer image processing offers potential for enhanced mammogram analysis.

Purpose of the Study:

• To develop an automated system for digital mammogram analysis.
• To classify regions of interest (ROIs) as masses or non-masses.
• To utilize texture features for improved breast cancer detection.

Main Methods:

• Image enhancement and ROI segmentation applied to digital mammograms.
• Texture feature extraction using Gray Level Co-occurrence Matrices (GLCM) at multiple directions.
• Classification of ROIs using a decision tree based on GLCM properties.

Main Results:

• GLCM analysis at 0°, 45°, 90°, and 135° provided significant texture information.
• The proposed method achieved a Receiver Operating Characteristic (ROC) curve area (Az) of 0.84 with Otsu's method.
• The system demonstrated good performance (Az = 0.8-0.9) in distinguishing between mass and non-mass tissues.

Conclusions:

• The developed automated system effectively aids in breast cancer diagnosis using digital mammograms.
• The GLCM-based texture analysis provides valuable information for mass detection.
• The system's simplicity and efficiency make it suitable for real-time automated diagnosis.