An Automatic Convolutional Neural Network Based Framework for Robust Classification of Breast Cancer Histopathological Images

Abstract:: Millions of deaths all over the world are caused by breast cancer every year. It has become the most common type of cancer in women. Early detection will help in better prognosis and increases the chance of survival. Automating the classification using Computer-Aided Diagnosis (CAD) systems can make the diagnosis less prone to errors. Multi class classification and Binary classification of breast cancer is a challenging problem. Convolutional neural network architectures extract specific feature descriptors from images, which cannot represent different types of breast cancer. This leads to false positives in classification, which is undesirable in disease diagnosis. The current paper presents an ensemble Convolutional neural network for multi class classification and Binary classification of breast cancer. The feature descriptors from each network are combined to produce the final classification. In this paper, histopathological images are taken from publicly available BreakHis dataset and classified between 8 classes. The proposed ensemble model can perform better when compared to the methods proposed in the literature. The results showed that the proposed model could be a viable approach for breast cancer classification.

Download Full-text

Prediction of Breast Cancer images Classification Using Bidirectional Long Short Term Memory and Two-Dimensional Convolutional Neural network

Transactions on Networks and Communications ◽

10.14738/tnc.94.10663 ◽

2021 ◽

Vol 9 (4) ◽

pp. 29-38

Author(s):

Oluwashola David Adeniji

Keyword(s):

Breast Cancer ◽

Neural Network ◽

Convolutional Neural Network ◽

Short Term Memory ◽

Screening Methods ◽

Cancer Treatments ◽

The Neural Network ◽

Histopathological Images ◽

Long Short Term Memory

Breast cancer is most prevalent among women around the world and Nigeria is no exception in this menace. The increased in survival rate is due to the dramatic advancement in the screening methods, early diagnosis, and discovery in cancer treatments. There is an improvement in different strategies of breast cancer classification. A model for training deep neural networks for classiﬁcation of breast cancer in histopathological images was developed in this study. However, this images are aﬀected by data unbalance with the support of active learning. The output of the neural network on unlabeled samples was used to calculate weighted information entropy. It is utilized as uncertainty score for automatic selecting both samples with high and low conﬁdence. A threshold that decays over iteration number is used to decide which high conﬁdence samples should be concatenated with manually labeled samples and then used inﬁne-tuning of convolutional neural network. The neural network was optionally trained using weighted cross-entropy loss to better cope with bias towards the majority class. The developed model was compared with the existing model. The accuracy level of 98.3% was achieved for the developed model while the existing model 93.97%. The accuracy gain of 4.33%. was achieved as performance in the prediction of breast cancer .

Download Full-text