Related Experiment Video
Updated: Jul 10, 2026
05:28
Clinical Imaging of Microwave Mammography
Published on: November 14, 2025
SVM based system for classification of microcalcifications in digital mammograms
Sukhwinder Singh1, Vinod Kumar, H K Verma
1Dept. of Comput. Sci. & Eng., Sant Longowal Inst. of Eng. & Technol., Longowal, India. sukhdalip@yahoo.com
Summary
This study introduces a computer-aided diagnosis (CAD) system using Support Vector Machines (SVM) to accurately classify clustered microcalcifications in mammograms, distinguishing between benign and malignant cases with high precision.
Area of Science:
- Medical Imaging
- Machine Learning
- Computer-Aided Diagnosis
Background:
- Clustered microcalcifications in mammograms are critical indicators for breast cancer detection.
- Accurate characterization of these microcalcifications is essential for timely diagnosis and treatment planning.
Purpose of the Study:
- To develop and evaluate a Support Vector Machine (SVM) based computer-aided diagnosis (CAD) system.
- To accurately characterize clustered microcalcifications in digitized mammograms as benign or malignant.
Main Methods:
- Image enhancement using the morphological enhancement (MORPHEN) method.
- Segmentation of potential microcalcification regions via edge detection and morphological operations.
- Feature extraction based on shape, texture, and statistical properties.
- Classification using SVM with Radial Basis Function (RBF) and polynomial kernels.
Main Results:
- The SVM with RBF kernel achieved an Area Under the Curve (A(z)) of 0.9803 with 97% accuracy.
- The SVM with a polynomial kernel achieved an A(z) of 0.9541 with 95% accuracy.
- The system demonstrated high performance in distinguishing benign from malignant microcalcification clusters.
Conclusions:
- The proposed SVM-based CAD system is effective for the characterization of clustered microcalcifications.
- The system shows significant potential for improving the accuracy and efficiency of breast cancer diagnosis from mammograms.
