A study of the suitability of autoencoders for preprocessing data in breast cancer experimentation

Abstract

Breast cancer is the most common cause of cancer death in women. Today, post-transcriptional protein products of the genes involved in breast cancer can be identified by immunohistochemistry. However, this method has problems arising from the intra-observer and inter-observer variability in the assessment of pathologic variables, which may result in misleading conclusions. Using an optimal selection of preprocessing techniques may help to reduce observer variability. Deep learning has emerged as a powerful technique for any tasks related to machine learning such as classification and regression. The aim of this work is to use autoencoders (neural networks commonly used to feed deep learning architectures) to improve the quality of the data for developing immunohistochemistry signatures with prognostic value in breast cancer. Our testing on data from 222 patients with invasive non-special type breast carcinoma shows that an automatic binarization of experimental data after autoencoding could outperform other classical preprocessing techniques (such as human-dependent or automatic binarization only) when applied to the prognosis of breast cancer by immunohistochemical signatures.

A Novel Ensemble Forecasting Algorithm based on Distributed Deep Learning Network

“…Recently, several big data scene methods based on MapReduce have been applied for traditional tasks. The parallel-based SVM algorithms have been applied in the field of high performance computing [9]. Several parallel KNN algorithms based on Spark have been proposed in [10].…”

Deep stacked sparse auto-encoders for prediction of post-operative survival expectancy in thoracic lung cancer surgery

“…They elicited nine decision rules from a thoracic surgery data set including 470 samples and 16 features for medical utilization, for predicting post-operative survival expectancy in lung cancer patients. Iraji (2017) has implemented a multi-layer architecture of sub-adaptive neuro-fuzzy inference system (MLA-ANFIS) with various combinations of multiple input features, regression, neural networks and an extreme learning machine (Macías-García et al, 2017) based on a thoracic surgery data set with 16 input features for prediction of 1-year post-operative survival expectancy in thoracic lung cancer surgery; ELM (wave kernel) has good performance.…”