Deep Learning Models Combining for Breast Cancer Histopathology Image Classification

Abstract

Breast cancer is one of the foremost reasons of death among women in the world. It has the largest mortality rate compared to the types of cancer accounting for 1.9 million per year in 2020. An early diagnosis may increase the survival rates. To this end, automating the analysis and the diagnosis allows to improve the accuracy and to reduce processing time. However, analyzing breast imagery's is non-trivial and may lead to experts' disagreements. In this research, we focus on breast cancer histopathological images acquired using the microscopic scan of breast tissues. We present combined two deep convolutional neural networks (DCNNs) to extract distinguished image features using transfer learning. The pre-trained Inception and the Xceptions models are used in parallel. Then, the feature maps are combined and reduced by dropout before being fed to the last fully connected layers for classification. We follow a sub-image classification then a whole image classification based on majority vote and maximum probability rules. Four tissue malignancy levels are considered: normal, benign, in situ carcinoma, and invasive carcinoma. The experimentations are performed to the Breast Cancer Histology (BACH) dataset. The overall accuracy for the sub-image classification is 97.29% and for the carcinoma cases the sensitivity achieved 99.58%. The whole image classification overall accuracy reaches 100% by majority vote and 95% by maximum probability fusion decision. The numerical results showed that our proposed approach outperforms the previous methods in terms of accuracy and sensitivity. The proposed design allows an extension to whole-slide histology images classification.

Deep Learning on Histopathology Images for Breast Cancer Classification: A Bibliometric Analysis

“…The analysis illustrates the social network structure that exists between individuals or organizations. Recently, the technological breakthrough in the CAD system has helped to improve the computational time of diagnosis and minimize the rate of misdiagnosis during image classification [ 93 , 94 ].…”

Breast Cancer Classification From Histopathological Images Using Resolution Adaptive Network

“…For the BACH 2018 dataset, the classification performance was evaluated for multiclass and binary classes based on two criteria: the image level and patch level [22,34]. Additionally, different metrics were used to estimate the effect of the ADSVM module on the classification system.…”

“…Regarding the analysis process for histopathological images, the data augmentation methods based on geometric transform were used to increase the quantity of data. The geometric transform includes rotation, flipping, and shifting which maintain the histopathological characteristic of the tissues without altering the classification results [33,34]. The classification performance of these methods (including rotation with different degrees, flipping, and shifting) was evaluated using the BreaKHis and the BACH 2018 datasets to select appropriate transform methods for data augmentation.…”

Robust ACO-Based Landmark Matching and Maxillofacial Anomalies Classification

“…A great deal of research has taken advantage of the flourishing deep learning technique in anomalies and diseases detection [8][9][10][11]. In 2017, Arik et al [12] use a convolutional neural network for a binary classification and achieved success detection rates of 75.58%, 75.37%, and 67.68% at a 2 mm range for three different datasets, respectively.…”