Date of Publication :15th March 2025

Abstract: This study suggests a hybrid deep learning model that blends the benefits of LSTM and CNN networks to accurately classify Wireless Capsule Endoscopy (WCE) pictures from the Kvasir dataset to meet the growing need for automated medical image analysis. Enhancing the model’s ability to handle intricate visual patterns, the CNN extracts spatial characteristics from the images, while the LSTM records temporal connections between these features. To reduce the impact of unbalanced data we use a weighted loss function and data augmentation. With an accuracy of 97.8%, experimental data show that our suggested model works superior to the state-of-the- art methods now in use. The accuracy and efficiencyWCE and gastrointestinal image analysis could be greatly increased by this research.

Reference :