Medical Image Classification Based on Fuzzy Support Vector Machines
Xingli Bai, Qian Xu · 2008
The paper present a novel method for medical image classification using fuzzy support vector machines (FSVM). In this method a membership degree is defined for each training sample, which can resolve the problem of unclassifiable regions in SVM. Experiments on images of mammography with different noise levels were conducted and results show that the proposed method is able to classify the breast cancer in the images of mammography with high precision. In application of this method the cost and time of computation can also be reduced.