Skip to main content Skip to navigation

Integrating Artificial Intelligence and Augmented Reality in Robotic Surgery: An Initial dVRK Study Using a Surgical Education Scenario

Project Overview

This document explores the application of generative AI and Augmented Reality (AR) in robotic surgery education, emphasizing a novel system that employs reinforcement learning to create 3D surgical guidance. The system aims to enhance training for novice surgeons by offering real-time AR overlays that illustrate guidance trajectories within a simulated surgical environment. The integration of AI and AR is presented as a promising solution to improve the accessibility of surgical education and lower training costs. Through preliminary experiments, the document assesses the feasibility of this innovative approach, demonstrating its potential to transform surgical training by providing immersive, interactive learning experiences that can significantly benefit aspiring surgeons. Overall, the findings suggest that such technologies could enhance skill acquisition and confidence among trainees, ultimately leading to better patient outcomes in surgical practices.

Key Applications

AI-powered AR system for robotic surgery education

Context: Surgical education for novice surgeons using the da Vinci Research Kit (dVRK)

Implementation: The system integrates reinforcement learning to generate surgical guidance trajectories, which are then visualized using AR in the dVRK console.

Outcomes: Successful demonstration of the system in a peg-transfer task, with an average success rate of 86.4%. The integration of AR enhances usability and user experience in surgical training.

Challenges: Challenges include ensuring the accuracy of the guidance trajectories and the precision of the AR overlays, as well as the potential for user error in coordinate calibration.

Implementation Barriers

Technical Barrier

Challenges in merging AI and AR technologies effectively for real-time surgical guidance.

Proposed Solutions: Developing robust algorithms and protocols to synchronize AI outputs with AR visualizations.

Project Team

Yonghao Long

Researcher

Jianfeng Cao

Researcher

Anton Deguet

Researcher

Russell H. Taylor

Researcher

Qi Dou

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Yonghao Long, Jianfeng Cao, Anton Deguet, Russell H. Taylor, Qi Dou

Source Publication: View Original PaperLink opens in a new window

Project Contact: Dr. Jianhua Yang

LLM Model Version: gpt-4o-mini-2024-07-18

Analysis Provider: Openai

Let us know you agree to cookies