Computer-aided Diagnosis of Breast Cancer Using Gaussian Mixture Cytological Image Segmentation

This paper presents an automatic computer system to breast cancer diagnosis. System was designed to distinguish benign from malignant tumors based on fine needle biopsy microscope images. Studies conducted focus on two different problems, the first concern the extraction of morphometric and colorimetric parameters of nuclei from cytological images and the other concentrate on breast cancer classification. In order to extract the nuclei features, segmentation procedure that integrates results of adaptive thresholding and Gaussian mixture clustering was implemented. Next, tumors were classified using four different classification methods: k-nearest neighbors, naive Bayes, decision trees and classifiers ensemble. Diagnostic accuracy obtained for conducted experiments varies according to different classification methods and fluctuates up to 98% for quasi optimal subset of features. All computational experiments were carried out using microscope images collected from 25 benign and 25 malignant lesions cases.