WCEbleedGen: A wireless capsule endoscopy dataset and its benchmarking for automatic bleeding classification, detection, and segmentation

TL;DR

This work tackles the lack of annotated datasets for automatic bleeding analysis in Wireless Capsule Endoscopy by introducing WCEbleedGen, a balanced dataset of 2,618 frames with bleeding and non-bleeding labels, binary masks, and bounding boxes. It provides a comprehensive benchmarking framework across nine classification models, three detection models, and three segmentation models, enabling multi-task learning in bleeding analysis. Results show high training performance but limited generalization for classification, moderate detection performance with YOLO variants, and strong segmentation performance (notably with LinkNet), underscoring the potential and need for further model tuning and data expansion. The publicly available dataset and code are designed to foster reproducible research and real-time, multi-task bleeding diagnosis in WCE contexts.

Abstract

Computer-based analysis of Wireless Capsule Endoscopy (WCE) is crucial. However, a medically annotated WCE dataset for training and evaluation of automatic classification, detection, and segmentation of bleeding and non-bleeding frames is currently lacking. The present work focused on development of a medically annotated WCE dataset called WCEbleedGen for automatic classification, detection, and segmentation of bleeding and non-bleeding frames. It comprises 2,618 WCE bleeding and non-bleeding frames which were collected from various internet resources and existing WCE datasets. A comprehensive benchmarking and evaluation of the developed dataset was done using nine classification-based, three detection-based, and three segmentation-based deep learning models. The dataset is of high-quality, is class-balanced and contains single and multiple bleeding sites. Overall, our standard benchmark results show that Visual Geometric Group (VGG) 19, You Only Look Once version 8 nano (YOLOv8n), and Link network (Linknet) performed best in automatic classification, detection, and segmentation-based evaluations, respectively. Automatic bleeding diagnosis is crucial for WCE video interpretations. This diverse dataset will aid in developing of real-time, multi-task learning-based innovative solutions for automatic bleeding diagnosis in WCE. The dataset and code are publicly available at https://zenodo.org/records/10156571 and https://github.com/misahub2023/Benchmarking-Codes-of-the-WCEBleedGen-dataset.

Figures (1)

• Figure 1: Sample bleeding and non-bleeding frames and their binary masks present in the WCEBleedGen dataset.