Optimizing capsule endoscopy detection: A deep learning approach with L-Softmax and laplacian-SGD

Capsule endoscopy has emerged as a non-invasive diagnostic tool for gastrointestinal diseases; however, efficient disease classification remains a challenge due to the inherent complexities of image analysis. Furthermore, the extensive time required for manual examination of capsule endoscopy images has led researchers and clinicians to seek timeefficient automated detection methods. This is where the profound advantages of deep learning (DL) become crucial. This research proposes a novel approach that combines L-Softmax with Laplacian Smoothing Stochastic Gradient Descent (LSSGD) within a ResNet architecture to enhance disease classification accuracy in capsule endoscopy images from the Kvasir dataset. The L-Softmax function is integrated into the DL framework, facilitating better class separation and feature representation. Additionally, LSSGD is employed to mitigate overfitting and enhance model generalization. Experimental results demonstrate that our methodology is stable and easy to utilize in capsule endoscopy.