Safety-aware robotic steering of a flexible endoscope for nasotracheal intubation. (April 2023)

• Record Type: Journal Article
• Authors: Deng, Zhen; Jiang, Peijie; Guo, Yuxin; Zhang, Shengzhan; Hu, Ying; Zheng, Xiaochun; He, Bingwei
• Is Part Of: Biomedical signal processing and control. Volume 82(2023)
• Publication Date: 2023-04
• Subjects: Robotic endoscopy -- Endoscopic image processing -- Depth estimation -- Visual feedback control -- Nasotracheal intubation; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2022.104504
• Languages: English
• ISSNs: 1746-8094

Abstract

Abstract: Automation in robotic endoscopy surgery is an important but challenging research topic. The flexible deformation of the endoscope makes it difficult for surgeons to steer it, especially in a narrow workspace. The safety of endoscope steering has been a major concern in a robotic endoscope. This study presents a safety-aware control framework to enable a robotic endotracheal intubation system (RNIS) to automatically steer the flexible endoscope during nasotracheal intubation (NTI). With the endoscopic image as feedback, we used the fusion of image intensity and optical flow measurements to detect the lumen center. A velocity-based orientation controller is designed to adjust the orientation of the endoscope tip to track the lumen center. To ensure surgical safety, the relative depth from the endoscope tip to the lumen surface is calculated from endoscopic images. With the relative depth as feedback, the feed motion of the endoscope is controlled to prevent the collision between the endoscope tip and its surrounding tissue. The position and orientation of the endoscope tip can be adjusted by the RNIS during NTI. Finally, extensive experiments based on a training manikin were performed. The results indicate that the proposed control framework enables the RNIS to successfully steer the flexible endoscope along the upper respiratory tract with high safety. The performance of lumen center detection is good according to the annotation of experienced users. Highlights: A safety-aware robot control framework for steering an endoscope is presented. Lumen center detection is performed to provide guidance for endoscope steering. Relative depth from the endoscope tip to the lumen surface is estimated.