The Application of Mixed Reality in Bronchoscopy Simulation Training: A Feasibility Study.

Abstract

Flexible bronchoscopy is an important invasive procedure used in the diagnosis and treatment of respiratory disorders. Training tools for bronchoscopy include virtual reality (VR) simulators and bronchial models comprised of molded silicone, both of which are associated with better learning outcomes and complement each other. Simulations using bronchial training models can provide a realistic experience using an actual bronchoscopy, but the information available for reference such as anatomical features and instructions during the simulation is insufficient. Microsoft HoloLens is a cordless Mixed Reality (MR) headset that enables users to view and operate holograms that do not exist in the real world, thereby providing remote advice and guided assistance, and is used in the medical field. Microsoft Dynamics 365 Guides is an MR application for HoloLens that allows users to create guides consisting of text displays, images, and three-dimensional (3D) holograms without prior programming expertise. We therefore created a training guide using Hololens2 and Dynamics 365 Guides to enable learners to efficiently learn bronchoscopy by themselves. First, we used an application in a personal computer to create instructions and schema for the three tasks: learning of bronchial anatomy, scope operation, and observation of the bronchial lumen. Next, after wearing the HoloLens2, the imported 3D image parts and operation guides were placed in virtual space (Figure 1(a)). Then, we simulated a bronchoscopy training using the MR guide. Ten physicians of varying experience performed the training program while wearing the Hololens2. Following the MR guide, the operator first learned about anatomy with a 3D hologram of the bronchus (Figure 1(b)), then, on how to operate the bronchoscope. Finally, they inserted the bronchoscope and observed the lumen using the training model (Figure 1(c) and (d)). The operator manipulated the holographic cards using instructions and images by eye or hand tracking. All participants underwent training once and were evaluated for the time required and with the questionnaire. Responses to the questionnaire are presented on a 5-point Likert scale (from 1 = strongly disagree to 5 = strongly agree). All participants completed the training, and the median time required was 10.7 minutes. The average score for visibility of the guide questionnaire ranged from 4.0 to 4.5 out of 5. They also averaged 4.7 and 4.6, respectively, on whether they would be willing to take the training again and recommend it to a colleague. The average score for wearing comfort of the device was 4.1, and eye strain and sickness were absent or mild for most participants. In the present study, participants responded favorably to using HoloLens2 to refer and manipulate guides during bronchoscopy training. Park et al. revealed that HoloLens is used for visualization of medical data, blood vessel search, targeting support for needles and endoscope support in medical and surgical aids and systems in their review. Suresh et al. reported in their review that augmented reality in surgical education was a feasible and effective adjunct to traditional training, and the Microsoft HoloLens had shown the most promising results among the several platforms. Heo et al. demonstrated that the use