Deep Convolutional Neural Network for the Prediction of Ovarian Cancer
S. Lakshmanan, P. Nagaraja, M. Mary Shanthi Rani, P. Shanmugavadivu · 2023
Ovarian cancer (OC) is the third most common gynaecological cancer globally, after cervical and uterine cancer, and it has a high mortality rate. It is the third leading cancer among women in India. The female reproductive system consists of two ovaries. OC is often detected when it has spread to the pelvis and abdomen, and it is very hard to treat at this stage. It can be successfully treated when the disease is confined to the ovary. As a result, understanding OC heterogeneity is critical for selecting different methods that help in predicting patients’ clinical outcomes. Pathologists currently rely on computer-aided diagnosis (CAD) for diagnosis and the common method of detection of heterogeneity is to identify different subtypes in OC. Recently, machine learning (ML) techniques have been widely used the OC prediction and classification. The proposed study aims to analyse three different pre-trained deep convolutional neural networks (DCNN) architectures, namely VGG16, VGG19 and Alexnet, to automatically detect, predict and classify the subtypes of OCs from histopathological images. The performance analysis of the study has been carried out with the parameters based on accuracy, F1 score, precision and recall.