Advancements in Machine Learning and Deep Learning for Ovarian Cancer Classification: A Comprehensive Survey

Ovarian cancer continues to represent a dangerous gynaecological malignancy because healthcare providers detect it in advanced stages and the survival rates remain unacceptably low. Modern detection and classification systems along with prognosis prediction improvement result from the implementation of artificial intelligence through machine learning (ML) and deep learning (DL). Various ML and DL methodologies including Support Vector Machines (SVM), Decision Trees (DT), ensemble models, and CNNs, RNNs together with hybrid AI approaches are applied to ovarian cancer classification according to this survey. The research acknowledges overlapping studies using histopathological images together with ultrasound scans and molecular data to reach high classification precision levels above 98%. The difficulty of finding sufficient data along with model overfitting and unclear interpretation and verification against clinical standards blocks the extensive adoption of these systems. The research field needs future work to expand available medical data while adopting transfer learning techniques, improving AI model explainability mechanisms, and applying federated learning systems for AI framework development. The diagnostic accuracy can receive additional optimization through the merging of multi-modal imagery systems and Internet of Medical Things technology alongside real-time cloud-based artificial intelligence solutions. By resolving these issues early detection methods will improve while personalized treatments become accessible which leads to decreased ovarian cancer mortality levels.