A Survey of Methods and Architectures for Semantic Segmentation and Classification for Ovarian Cancer Detection Using EfficientNetB0 with FPN and Causal Dilated Convolutional Neural Networks

Abstract

Ovarian cancer is one of the leading causes of cancer-related mortality among women due to its asymptomatic nature in early stages and the lack of effective screening techniques. Medical imaging modalities such as ultrasound, MRI, and CT scans are commonly used for diagnosis; however, manual interpretation is time-consuming and prone to variability. Recent advancements in deep learning, particularly convolutional neural networks (CNNs), have significantly enhanced the accuracy and efficiency of automated detection systems. This survey provides a comprehensive overview of semantic segmentation and classification techniques for ovarian cancer detection, focusing on architectures such as EfficientNetB0, Feature Pyramid Networks (FPN), and causal dilated convolutional neural networks. EfficientNetB0 offers an optimal balance between accuracy and computational efficiency, while FPN improves multi-scale feature representation for precise tumor localization. Semantic segmentation models like U-Net variants effectively delineate tumor boundaries, and classification models distinguish between benign and malignant cases. These approaches achieve high diagnostic performance, often comparable to expert radiologists. Despite these advancements, challenges such as limited datasets, model generalization, and interpretability persist, highlighting the need for explainable, scalable, and clinically applicable AI solutions.