Prediction of IDH1 Gene expression in Glioblastoma Using Machine Learning Techniques

The most common malignant primary elementary brain tumor is Glioblastoma, accounting for more than 50% of all invasive primary brain tumors. Isocitrate dehydrogenase 1 (IDH1) is an independent beneficial prognosis characteristic for glioblastoma (GBM). Recently, the gene mutations of IDH1 were found in 20, 661 genes that code for proteins involved in glioblastomas due to IDH1 mutation prediction can significantly lead to secondary glioblastoma diagnosis and glioblastoma prediction. An accurate prediction of IDH1 mutation status assists in establishing the precise treatment method that helps to minimize the significant risk from GBM. This paper investigates the IDH1 gene expression in glioblastoma disease detection. We are exploiting prevalent machine learning classification algorithms, namely, Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), Neural Network (NN), K-Nearest Neighbors (K-NN), and Decision Tree (DT). Using performance metrics, we compare and evaluate the outcomes of these machine learning approaches. In the experiment, we get the highest accuracy (98.8%) for the Support Vector Machine classifier for IDH1 gene expression classification.