Capsule Endoscopy

Capsule endoscopy (CE) is a minimally invasive technique for visualizing the gastrointestinal tract, primarily aimed at improving diagnosis of conditions like gastrointestinal bleeding and cancer. Current research heavily emphasizes the application of artificial intelligence, particularly convolutional neural networks (CNNs) and transformer models, often combined, to automate image analysis, improve diagnostic accuracy, and reduce the time-consuming manual review of large image datasets. This work addresses challenges such as uneven illumination, highlight removal, and class imbalance in CE images, ultimately aiming to enhance the efficiency and effectiveness of this crucial diagnostic tool.