Enhancing Breast Cancer Prediction through Deep Learning and Comparative Analysis of Gene Expression and DNA Methylation Data using Convolutional Neural Networks

Abstract

Recent advances in the production of statistics have resulted in an exponential increase in the number of facts, ushering in a whole new era dominated by very large facts. Conventional machine-learning algorithms are unable to handle the most recent aspects of huge data. This is a fact.  In order to make an accurate prognosis of breast cancer, researchers employ and evaluate three distinct computer programmes called Support Vector Machine (SVM), Random Forest (RF), and Decision Tree (DT). Within the context of huge statistics, we explore the question of how breast cancer may be predicted in this particular research. Gene expression and DNA methylation are both taken into consideration as part of the analysis (GE and DM, respectively). The purpose of the work that we are doing is to increase the capacity of the Deep Learning algorithms that are now being used for typing by applying each dataset individually and together. As a result of this decision, the platform of choice is MATLAB. In the process of breast cancer prediction, the Convolutional Neural Network (CNN) algorithm is used. Comparisons of GE, DM, and GE and DM are carried out with the help of this method. The results of the CNN algorithm are compared to those of the RF algorithm. According to findings of the experiments, the scaled system that was presented works better than the other classifiers. This is due to the fact that using the GE dataset; it acquired the best accuracy at the lowest cost.