International Session（Symposium）9(JGES・JSGE・JSGS・JSGCS)
Sat. November 7th   9:00 - 11:30   Room 11: Portopia Hotel South Wing Topaz
IS-S9-8_E
 Marrying AI with robot to transform GI endoscopy
Khek Yu Ho
National University of Singapore
Robotic endoscopic platforms, such as the Master and Slave TransEndoluminal Robot (MASTER), provide unique features for its users. With 2 robotic arms coming out from the distal ends of the working channels of the endoscope, the operator can control the robotic arms’ movement from a master console. The robotic arms provide triangulation, haptic feedback, and allow the operator to manipulate the 2 arms intuitively. The Master System has been validated in human trials for endoscopic submucosal dissection of early GIT neoplasia.

However, for the robotic endoscope to be adopted by mainstream clinical practice, its continued relevance to practitioners must be ensured. One such application is the adaptation of the robotic endoscopy for the performance of NOTES. We have recently developed a new capability of the Master System, allowing it to perform suturing and knot tying. Closure using the endoscopic robotic suturing method is expected to be as strong as a surgical closure.

Although the robotic solution can help improve the technical skills of the procedurist, it does not enhance the procedurist’s clinical decision-making process. This challenge can be overcome by adapting an artificial intelligence-based deep learning system to the robotic endoscope. We have developed the world’s one-of-a-kind In-Vivo Molecular Diagnostic System, which can be used to make realtime diagnosis of GI cancer. Its significance lies in the ability of the technology to help procedurists decide the resection margin during surgical procedures.
Index Term 1: Robotic endoscopy
Index Term 2: Artificial intelligence
Page Top