Atul Kumar1, Kai-Che Liu1, Benjamin Niay2, Shih-Wei Huang1, Yen-Yu Wang3, Yi-Chun Lee4. 1Show Chwan Memorial Hospital, Changhua, 2Télécom Physique Strasbourg, France, 3National Changhua University of Education, Changhua, 4Chang Bing Show Chwan Memorial Hospital, Changhua
Objective: Augmented reality (AR) has been recently applied in endoscopic surgery [1, 2]. In AR a virtual object is superimposed on a real scene to enhance the information from the real scene. In AR applied to endoscopic surgery, a 3D virtual model of human organs reconstructed from CT or MRI images are superimposed on the endoscopic video. However, to our knowledge, there is no web based platform which can provide AR for surgical use. This work presents a web application which can display a superimposed view of an endsocope live video and a 3D virtual model of human torso phantom.
Method: The web application is written using HTML and JavaScript programming languages using Three.js [3], a 3D animation library for JavaScript. The application can use any camera attached to the computer as video input and display the video as textured background of the 3D virtual environment displayed on the web page. The 3D virtual model of an object is further added to the virtual environment to visualize a combined view of real and virtual objects.
Results: A 3D virtual model (Fig. 1) of human torso phantom (IOUSFAN, Kyoto Kagaku Co. Ltd., Kyoto, Japan), was reconstructed from its CT scan. An endoscope system of Karl Storz R was used for video input (Fig. 2). The application was run in several kinds of web browser(such as mozilla firefox and google chrome) to visualize the 3D model, endoscope video and their combined view (Fig. 3). The endoscope video can be displayed in real-time.
Conclusion: A web application to visualize superimposed endoscope video and 3D virtual model was developed. This is an initial step towards developing a web application for augmented reality which can be used by clinicians. The current application would be further developed to include a image based camera tracking system. The complete system would further be analyzed for its tracking accuracy.
Fig. 1 A 3D model of phantom reconstructed from CT images displayed on web page
Fig. 2 An endoscope video of the phantom on web page
Fig. 3 A combined view of endoscope video and the 3D model of the phantom on web page
References
1. Kang, Xin, et al: Stereoscopic augmented reality for laparoscopic surgery. Surgical endoscopy 28.7 (2014): 2227-2235.
2. Wang, Rong, et al: Visualization Techniques for Augmented Reality in Endoscopic Surgery. International Conference on Medical Imaging and Virtual Reality. Springer International Publishing, 2016.
3. Dirksen, Jos: Learning Three. js: the JavaScript 3D library for WebGL. Packt Publishing Ltd, (2013).
Presented at the SAGES 2017 Annual Meeting in Houston, TX.
Abstract ID: 91099
Program Number: ETP885
Presentation Session: Emerging Technology iPoster Session (Non CME)
Presentation Type: Poster