An Efficient Approach for Automated Mass Segmentation and Classification in Mammograms

Area of Science:

• Medical Imaging
• Computer-Aided Diagnosis
• Machine Learning in Healthcare

Background:

• Breast cancer is a leading cause of death for women globally.
• Early detection and accurate diagnosis are crucial for effective breast cancer treatment.
• Existing diagnostic methods require improvement in precision and efficiency.

Purpose of the Study:

• To develop a novel automated segmentation and classification method for mammograms.
• To enhance the accuracy of breast cancer diagnosis using artificial intelligence.
• To create a system that assists physicians in clinical decision-making.

Main Methods:

• Region of Interests (ROIs) extraction using chain codes and Rough Set (RS) method enhancement.
• Mass segmentation from ROIs utilizing an improved Vector Field Convolution (VFC) snake.
• Feature extraction (32 dimensions) and classification using Random Forest, compared against SVM, GA-SVM, PSO-SVM, and Decision Tree.

Main Results:

• The proposed method achieved a high accuracy of 97.73% on the DDSM database.
• Matthew's Correlation Coefficient (MCC) reached 0.8668 (DDSM) and 0.8652 (combined databases).
• The Random Forest classifier outperformed other state-of-the-art techniques in this study.

Conclusions:

• The novel automated method demonstrates superior performance in breast cancer mammogram analysis.
• The approach shows significant potential for integration into Computer-Aided Diagnosis (CAD) systems.
• Accurate segmentation and classification can aid physicians, leading to improved patient outcomes.