Ovarian Cancer Detection using Microarray gene data and Deep Learning Techniques

Ovarian cancer, is one of the most common gynecologic problem, that poses a severe threat to women’s health. Since there are tens of thousands of genes in the microarray data for ovarian cancer, it is extremely Challenge to detect the disease. Reducing computational complexity by selecting the most relevant genes is crucial due to the limited sample size and high dimensionality of these datasets. In this research, the ANOVA method is applied for initial gene selection during the preprocessing stage. Next, performing statistical parameter analysis to gain insights into the dataset. Feature selection and dimensionality reduction are conducted using a Genetic Algorithm to optimize gene selection. For prediction, a Graph Convolutional Network (GCN) is employed to classify normal and abnormal samples. Finally, the performance of the model is evaluated through comprehensive metrics analysis. This workflow is designed to improve the accuracy and efficiency of ovarian cancer detection by focusing on the most informative genes and utilizing advanced Deep learning techniques. This proposed methodology is achieved the accuracy with 85.93% demonstrating that the integration of GA for feature optimization with GCN for classification significantly improves detection performance compared to traditional methods.