A GLCM based Feature Extraction in Mammogram Images using Machine Learning Algorithms
B. Naga Jagadesh, L. Kanya Kumari · International Journal of Current Research and Review · 2021
Introduction: Most Indian women are suffering from Breast Cancer.The simple and efficient screening used for Breast Cancer (BC) is Mammograms.Mammogram images are used to detect BC in the early stages.Objective: The main objective of our research is to detect the BC in early stages using Gray Level Co-occurrence Matrix (GLCM) with Machine Learning Algorithms.Methods: Our proposed system is a two-step process which includes feature extraction and classification.Features are extracted from the Mammographic Image Analysis Society (MIAS) database by using a texture-based descriptor called GLCM.These features are passed to classifiers called K-Nearest Neighbor (KNN), Random Forest (RF) and Gradient Boosting by considering 30% as testing data size. Results:The experiments are done as follows: GLCM+RF, GLCM+KNN and GLCM+ Gradient Boosting and the performance of these classifiers are calculated by finding accuracy metric. Conclusion:The conclusion is that GLCM features with KNN classifier give better results than other classifiers.