ABSTRACT
Analysing histopathological images to diagnose breast cancer plays a crucial role in the patients’ development, and deep learning techniques can analyse a set of parameters from images utilizing deep convolutional networks. Using transfer learning techniques, we modified shuffle network architectures for the multi-class image classification of histopathological images of breast cancer. All our results are used to show that transfer learning provides the best performance analysis of histopathological images of breast cancer. The average accuracy of all classes of breast tumor cells is 95±% using MATLAB. In this chapter, we have proposed our method of using pre-trained networks in the transfer learning process using MATLAB on histopathological images of breast cancer. All results shown in this chapter are on an augmented dataset in multi-class classification. Future researchers could use different networks with other training options and different parameters.
