Detection of breast cancer from histopathology image and classifying benign and malignant state using fuzzy logic

Abstract

Breast cancer is one of the major public health problem for women throughout the world. It has two states, known as benign and malignant. Benign state is slow growing, rarely spread to other parts of body and have well-defined borders. On the other hand, Malignant state has tendency to grow faster and it is life threatening. So, classification of this two state is crucial for proper diagnosis of a breast cancer patient. In this paper, we have introduced a new pipeline for breast cancer cell detection and feature extraction using an open source image analysis software named CellProfiler. We proposed an algorithm based on fuzzy inference system for classification of the benign and malignant state. Comparison using well known performance parameters such as accuracy, sensitivity and specificity shows that our proposed approach performs better than the Artificial Neural Network (ANN) and Support Vector Machine (SVM) based classification. The sensitivity, specificity, and accuracy of the proposed method is 95.6%, 90.63%, and 94.26% respectively.