Artificial Intelligence-Based Polyp Detection And Measurement Add-On System For Endoscopy

The intelligent medical assistive system has emerged as a topic of considerable interest within the research community in recent years. The implementation of such a system has the potential to significantly reduce the workload of medical personnel while simultaneously enhancing the quality of medical treatment. The automated detection of endoscopic images through computer vision technology represents a significant aspect of the development of intelligent medical assistive systems. In this study, we propose the integration of artificial intelligence-based automated detection and measurement technology into a medical assistive system. The system is capable of detecting polyps in endoscopic images and measuring their sizes simultaneously, as well as calculating the distance to any region of interest in endoscopic images. The detection and measurement performance of the proposed system has been evaluated using self-collected real-world endoscopy images. To demonstrate the clinical unmet need that the proposed system addresses, it was deployed to an edge computing device and integrated with actual endoscopic equipment, which confirmed the effectiveness of the proposed system.