Abstract
Ovarian Cancer is a malignancy that develops from ovarian cells and is frequently characterized by aberrant cell proliferation that leads to the creation of tumors within the ovaries. The high death rate and often delayed discovery of Ovarian Cancer make it a serious healthcare concern. Due to the annual 207,000 fatalities and 314,000 new cases worldwide, Ovarian Cancer poses a serious threat to public health, making quick and precise detection and classification techniques more essential. This work discusses the importance of Ovarian Cancer diagnosis and presents a new model for Ovarian Cancer classification. It also showcases a comparative analysis with other state-of-the-art models for Ovarian Cancer. Using an Ovarian Cancer image dataset which has data samples named Clear Cell, Endometri, Mucinous, Serous, and Non-Cancerous, it compares the proposed OvCan-FIND model to a wide range of CNN-based architectures, such as Inception V3, different EfficientNet variants, ResNet152V2, MobileNet, MobileNetV2, VGG16, VGG19, and Xception. The study examines the most recent Ovarian Cancer classification algorithms in this context to increase prognosis and diagnostic accuracy; our proposed OvCan-FIND model outperforms base models with an exceptional accuracy of 99.74%. This model presents significant prospects for enhancing ovarian cancer early identification and diagnosis, which will ultimately enhance patient outcomes.