An Optimized Deep Convolutional Neural Network Approach to Detecting Ovarian Cancer

The ovaries are essential for the functioning of the female genital system. The significance for these limited glands lies in their role of synthesizing female sex hormones or female genital cells. The paired, endocrine, ovoid glands are located on either side of the womb and are linked to the ovarian ligament. Ovarian cancer can result from various sources, but it can be identified by multiple approaches, including the utilization of a technique called convolutional neural network. This study investigates the utilization of CNN for the classification of ovarian cancer tumors, as well as other approaches to address this issue. This study examines and compares different machine learning algorithms, such as K Nearest Neighbor (KNN) and the Support Vector Machine, to detect ovarian cancer. Upon evaluating various techniques for detecting cancer, it quickly became evident that a certain Deep Learning methodology, specifically the Deep CNN, yielded the most favorable results. The application of Deep Convolutional Neural Networks (DCNN) has led to a significant improvement in the accuracy of predicting cancer of the ovary and its subtypes, achieving a remarkable accuracy of 96%. This surpasses the accuracy of 83.3% attained by Support Vector Machines (SVM) and K-Nearest Neighbors (KNN) algorithms.