Integrating Deep Feature Extraction and Hybrid ResNet-DenseNet Model for Multi-Class Abnormality Detection in Endoscopic Images
This work addresses the diagnostic burden on gastroenterologists by providing an AI tool for multi-class abnormality detection in endoscopic images, though it is incremental in nature.
The paper tackled automated detection of ten gastrointestinal abnormalities in endoscopic images using a deep learning ensemble, achieving 94% overall accuracy with precision ranging from 0.56 to 1.00 and recall up to 98%.
This paper presents a deep learning framework for the multi-class classification of gastrointestinal abnormalities in Video Capsule Endoscopy (VCE) frames. The aim is to automate the identification of ten GI abnormality classes, including angioectasia, bleeding, and ulcers, thereby reducing the diagnostic burden on gastroenterologists. Utilizing an ensemble of DenseNet and ResNet architectures, the proposed model achieves an overall accuracy of 94\% across a well-structured dataset. Precision scores range from 0.56 for erythema to 1.00 for worms, with recall rates peaking at 98% for normal findings. This study emphasizes the importance of robust data preprocessing techniques, including normalization and augmentation, in enhancing model performance. The contributions of this work lie in developing an effective AI-driven tool that streamlines the diagnostic process in gastroenterology, ultimately improving patient care and clinical outcomes.