Optimal gene therapy network: Enhancing cancer classification through advanced AI-driven gene expression analysis

Abstract

Gene therapy is an advanced medical approach that aims to find solutions for various cancers by identifying optimal gene expressions. In this context, computer-aided detection of gene expressions becomes a research challenge, where artificial intelligence methods are employed to classify cancer types. However, traditional machine learning models must be improved for accurately classifying cancers, leading to unsatisfactory quantitative performance. Therefore, this work implemented the optimal gene therapy network (OGT-Net) for identifying the different types of cancers from the gene expression sequences. Initially, the dataset pre-processing operation normalizes the dataset, which maintains the uniform nature of all records in the dataset. Then, the light gradient boosting model (LGBM) extracts the correlated features from the pre-processed dataset, which contains the relationship among the pre-processed gene expression data. In addition, interrupt-based Harris Hawk optimization (IHHO) extracts the optimal features from LGBM data, decreasing the total number of features by removing redundant gene sequences. Then, a customized deep learning convolution neural network (DLCNN) is used to categorize diseases using gene expression datasets based on lymphography, colon, lung, ovarian, and prostate cancers. The simulation results reveal that the proposed OGT-Net improved performance on various datasets compared to existing approaches, with an average accuracy of 91.128 %, precision of 90.836 %, recall of 91.25 %, and F1-score of 90.7 %.