Breast Cancer Case Identification Based on Deep Learning and Bioinformatics Analysis

Understanding the molecular mechanism of breast cancer (BC) can provide an in-depth understanding of BC pathology. The purpose of this article is to use gene expression profiles from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) to classify breast cancer samples and normal samples. This study first selected the genes that are most relevant to cancer through weighted gene co-expression network analysis (WGCNA) and differential expression analysis (DEA), then used the protein-protein interaction (PPI) network to screen 23 core genes, and finally used the support vector machine (SVM), artificial neural network (ANN), convolutional neural network (CNN)-LeNet and CNN-AlexNet processes to determine the expression levels of 23 core genes. The ANN model had the best effect. Its average accuracy rate was 97.36%, the F1 value was 0.897, the precision rate was 93.6%, and the recall rate was 95.2%. In summary, this process effectively classifies cancer samples and normal samples and provides reasonable new ideas for the early diagnosis of cancer in the future.