Next Generation Robotic System for Supervised-Autonomous Bowel Anastomosis

Johns Hopkins University, Baltimore MD

Investigators

Linked publications, trials & patents

Abstract

Project Summary: The goal of this project is to establish a new medical robot and surgical paradigm to perform clinical colonic anastomosis with shared control. We will develop and evaluate novel miniaturized functional 3D imaging, artificial intelligence (AI) based real time soft tissue tracking, and fail-safe autonomous control strategies that enable first in human autonomous robotic soft tissue surgery. The long-term goal of this proposal is to reduce the complication rates and improve functional outcomes of bowel anastomoses and other soft tissue surgeries independent of surgeon’s experience. Anastomoses are critical and challenging procedures performed over a million times per year in the US. However, up to 19% of gastrointestinal anastomoses are complicated by leakage, strictures, and stenosis, and anastomotic complications significantly increase patient mortality up to ten times. Recent key advances in surgery such as progressively more minimally invasive surgery (MIS) and robot- assisted surgery (RAS) using a tele-operated da Vinci robot have not addressed the critical factors influencing anastomotic outcome, and the overall anastomotic complication rates remain unchanged. Autonomous robotic surgery systems have the potential to significantly improve efficiency, safety, and consistency over current tele- operated RAS. Autonomous robotic soft tissue surgery is yet to be demonstrated in humans due to a lack of (a) accurate 3D endoscopic vision, (b) robust markerless soft tissue tracking, and (c) fail safe autonomous control. To address the clinical need for autonomous robotic suturing platforms, we developed a supervised autonomous robotic system for guiding reconstructive bowel anastomosis. We propose to conduct first-in-human pilot studies of an autonomous robotic suturing platform. Because complications from bowel anastomoses can be catastrophic, we implement a stepwise gradual increase of complexity starting with open overstitching of the closure of a common enterotomy and building up to laparoscopic colectomy to ensure patient safety. We hypothesize that autonomous robotic soft tissue surgery is feasible in humans and will perform the task of laparoscopic colectomy with better accuracy and efficiency over current tele-robotic RAS technology. The following four specific aims will enable the development of this technology and demonstrate clinical feasibility. We will develop miniaturized functional 3D imaging and tissue health sensing (Aim 1), AI based markerless tissue tracking (Aim 2), fail-safe autonomous control strategies (Aim 3), and demonstrate first in human autonomous soft tissue surgery in the form of common enterotomy closures (Aim 4). Beyond intestinal anastomosis, adoption of this approach could be beneficial in all soft tissue MIS and RAS tasks requiring precision and maneuverability, including complex pediatric, urological, and cardiac surgery.