Wireless Capsule Endoscopy (WCE) enables non-invasive visualization of the gastrointestinal (GI) tract, but manual analysis of the thousands of images produced per examination is time consuming and error-prone. Deep learning approaches, including Convolutional Neural Networks (CNNs), transfer learning, and hybrid models, have shown promise in automating bleeding detection and segmentation in WCE images. This paper reviews recent advances in these techniques, compares their performance metrics, and discusses challenges such as limited annotated datasets, class imbalance, and model generalization. Future directions, including attention-based models, data augmentation with Generative Adversarial Networks (GANs), and explainable AI, are highlighted to guide further research. The insights provided aim to improve clinical workflows and enhance diagnostic efficiency for gastrointestinal bleeding detection.
Wireless Capsule Endoscopy, Deep Learning, Bleeding Detection, Convolutional Neural Networks, Transfer Learning, Hybrid Models, Gastrointestinal Diagnostics.