HARNESSING THE POWER OF ARTIFICIAL INTELLIGENCE FOR ACCURATE OVARIAN CANCER CLASSIFICATION

Abstract

Ovarian cancer’s high death rates, which are mostly brought on by late-stage detection, make it a major global health concern. Early detection is needed to improve survival and for effective treatment. This study explores how artificial intelligence (AI) might improve the identification of ovarian cancer, which is in line with Sustainable Development Goal 3 of the UN, which aims to promote health and well-being. We propose a combination of machine learning and transfer learning techniques, based on an open-source medical imaging dataset specifically for ovarian cancer. To extract intricate and hierarchical information from medical photos the study uses the pre-trained VVG16 deep learning model, The XGBoost algorithm, a potent ensemble learning method renowned for its excellent performance and efficiency was then used to classify these characteristics. The suggested approach showed robustness and reliability with an outstanding accuracy of 98.38% on the dataset in detecting ovarian cancer. The study points that there is great potential in the promise of AI-driven solutions and transfer learning in medical diagnostics, especially for diseases like ovarian cancer where early and accurate detection can greatly enhance results. The insights underscore the potential of AI to revolutionize healthcare and improve patient outcomes by providing timely treatments, potentially leading to the development of scalable, effective, and accurate diagnostic tools.