Indian Journal of Science and Technology

Abstract

Background/Objectives: Breast cancer, is a highly diverse disease and women commonly witnesses this. This paper inducts an intelligent classification method for identifying the breast cancer from the images attained as benign and malignant. This elementary test mode supports in recognizing breast cancer at initial period and this initial stage discovery would support in recovering more number of women from this serious disease. Statistical Analysis: An intelligent system is projected which practices Artificial Neural Networks (ANN) that contributes the user; choices to analyze, sense and quantity the cancer. To attain exact outcome, the images acquired by different medical imaging modalities must be assessed using machine learning algorithms. A variety of features are extracted from the image for detecting and diagnosing nonthreatening and malicious tumour forms in digital mammograms. Findings: A number of features are extracted and established a blend of three or four features, such as entropy, standard deviation, area etc to discriminate a benign tumour from the malignant one. The accuracy using Classification using FFNN algorithm is 99%.
Keywords: Artificial Neural Networks, Breast Cancer, Feature Extraction, Mammogram, Wavelet Transform