Two-phase deep convolutional neural network for reducing class skewness in histopathological images based breast cancer detection. (1st June 2017)

• Record Type: Journal Article
• Authors: Wahab, Noorul; Khan, Asifullah; Lee, Yeon Soo
• Is Part Of: Computers in biology and medicine. Volume 85(2017)
• Page Start: 86
• Page End: 97
• Publication Date: 2017-06-01
• Subjects: Convolutional neural networks -- Mitosis count -- Deep learning -- Histopathology -- Class imbalance -- Breast cancer; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2017.04.012
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: Different types of breast cancer are affecting lives of women across the world. Common types include Ductal carcinoma in situ (DCIS), Invasive ductal carcinoma (IDC), Tubular carcinoma, Medullary carcinoma, and Invasive lobular carcinoma (ILC). While detecting cancer, one important factor is mitotic count – showing how rapidly the cells are dividing. But the class imbalance problem, due to the small number of mitotic nuclei in comparison to the overwhelming number of non-mitotic nuclei, affects the performance of classification models. This work presents a two-phase model to mitigate the class biasness issue while classifying mitotic and non-mitotic nuclei in breast cancer histopathology images through a deep convolutional neural network (CNN). First, nuclei are segmented out using blue ratio and global binary thresholding. In Phase-1 a CNN is then trained on the segmented out 80×80 pixel patches based on a standard dataset. Hard non-mitotic examples are identified and augmented; mitotic examples are oversampled by rotation and flipping; whereas non-mitotic examples are undersampled by blue ratio histogram based k-means clustering. Based on this information from Phase-1, the dataset is modified for Phase-2 in order to reduce the effects of class imbalance. The proposed CNN architecture and data balancing technique yielded an F-measure of 0.79, and outperformed all the methods relying on specific handcrafted features, as well as those using a combination of handcrafted and CNN-generated features. Abstract : Graphical abstract: Abstract : Highlights: Performance of Convolutional Neural Network (CNN) suffers from data imbalance. A CNN with hybrid (data and model based) balancing is proposed to detect mitoses. Clustering based undersampling outperforms random undersampling. CNN-generated features can help avoid hand engineering features.