Breast cancer detection using an ensemble deep learning method. (September 2021)

• Record Type: Journal Article
• Authors: Das, Abhishek; Narayan Mohanty, Mihir; Kumar Mallick, Pradeep; Tiwari, Prayag; Muhammad, Khan; Zhu, Hongyin
• Is Part Of: Biomedical signal processing and control. Volume 70(2021)
• Publication Date: 2021-09
• Subjects: Convolutional neural networks -- Multilayer perceptron -- Empirical wavelet transform -- Variational mode decomposition -- Stacking ensemble -- Breast cancer -- Invasive ductal carcinoma; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2021.103009
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: Ensemble learning based model is designed for classification. One of the three CNNs is trained with the images in the dataset directly. One of the other two CNN is trained with the decomposed images generated by 1D Empirical Wavelet Transform applied to rows and columns and transformation to form two dimensional data. The third CNN model is also trained using VMD in the similar manner. The decomposed forms of original dataset are considered so that the models can be trained in molecular level. The first stage classification outcomes are termed as Meta data that is used to train the second stage classifier MLP. In parallel approaches, the model training process can reducethe dependence in the premise of ensuring the model convergence. It ensures that the convolution neuralnetwork framework is more adaptive to different system environments and can achieve the high speed as compared to the traditional single models with higher accuracy. To reduce the variance and generalization error, it is required to train multiple models instead of a single model and combine the predictions from these models. The final classification accuracy due to the proposed stacked ensemble is 98.08% that shows the effectiveness of considering such a technique. Abstract: In this work, the effectiveness of the deep learning model is applied for one-dimensional data when converted to images. This work is based on the effective conversion of one-dimensional data to images and designing a stacked ensemble deep learning model that can increase the performance of classification accuracy in comparison to single models. Breast cancer detection from gene expression dataset and breast histopathology images is considered using the proposed ensemble model. The gene expression data is one-dimensional. Using the t-Distributed Stochastic Neighbor embedding technique and Convex Hull algorithm the one-dimensional data is converted to an image. Existing methods are using the datasets directly for training and classification, whereas the proposed method uses the dataset as well as the decomposed forms of the same for improving the performance. It involves two-stage classification. The first stage consists of three Convolutional Neural Networks as the base classifiers. Empirical Wavelet Transform and Variational Mode Decomposition are the two methods used to decompose the dataset so that the models can be trained at the molecular level, making our model robust in comparison to state-of-the-art methods. The first stage classification outcomes are used to train the second stage classifier "Multilayer Perceptron". The gene expression dataset collected from Mendeley and is used for the generation of two-dimensional synthetic datasets. The synthetic datasets and breast histopathology image datasets are used for the training and validation of the proposedmodel. The improved results obtained in this work show the effectiveness of our method