Survey of AI-driven techniques for ovarian cancer detection: state-of-the-art methods and open challenges

Early detection is crucial for increasing the chance of survival in Ovarian Cancer (OC), as it is a very challenging illness to treat that often leads to death. Ultrasound (UT), Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) scans are 3 types of medical imaging techniques that play a crucial role in the diagnosis and treatment of OC. Furthermore, the definitive diagnosis is established through the examination by histological images. However, accurately interpreting medical images requires a substantial amount of information and expertise. Artificial Intelligence (AI), Deep learning (DL), and Machine learning (ML) have demonstrated promising applications and outcomes in medical image analysis, biomarker discovery, and therapy planning. This paper reviews the latest developments in OC detection using ML and DL techniques, focusing on significant progress made from 2019 to June 2024. The enhanced performance of DL algorithms on complicated medical images enables more precise and efficient diagnosis and treatment planning. Furthermore, we offer a concise explanation of the common evaluation standards used. Despite the notable progress in DL-based technologies, numerous problems remain. Consequently, we conducted a thorough analysis of these approaches to create a comprehensive list of research challenges that remain unresolved. These problems could potentially create novel areas of study for researchers across the globe.