Robotic system improves surgical removal of sticky small fragments
AGRO-SUVIDE: Agentic Robotics for Surgical Viscoelastic Debridement
Robotics
Summary
Removing tiny particles stuck to soft, sticky surfaces during surgery can be tiring and tricky for surgeons. The authors created a robotic system called AGRO-SUVIDE that learns from watching an expert surgeon just once. It breaks down the task into smaller skills that it can practice and improve on its own. When tested on a robotic surgery platform, it successfully removed most fragments without human help and could handle new cases with good success.
What this means in practice
- •For surgical robotics engineers: Build robotic surgery assistants that autonomously clear small debris from sticky tissue surfaces using learned skills from expert demonstrations.
- •For industrial automation teams: Develop robots that repeatedly remove small particles attached to soft materials in manufacturing using modular skill learning.
Authors
Shutong Jin, Ziyang Chen, Preethi Satish, Medow Shen, Gary Guthart, Florian T. Pokorny, Ken Goldberg
Abstract
Augmented dexterity has the potential to reduce the fatigue experienced by surgeons during repetitive surgical tasks. In this paper, we propose the first AGentic RObotics framework for SUrgical VIscoelastic DEbridement (AGRO-SUVIDE), the repeated removal of small fragments attached to a viscoelastic substrate. Leveraging the self-improving and coding capability of agents, AGRO-SUVIDE adopts a modular framework. Specifically, the demonstration analysis module automatically identifies recurring skills from a single expert demonstration, using both visual and kinematic information. The construction module then builds each skill, either as a procedural model-based skill the agent codes against a scaffolded library or as a model-free policy-based skill. At runtime, the monitoring module composes the skills into a loop-style graph sized to the number of fragments it observes, then verifies pre- and post-conditions of each skill to decide whether to advance or retry. We evaluate AGRO-SUVIDE through 340 physical trials on the da Vinci Research Kit (dVRK). AGRO-SUVIDE achieves an average single-fragment removal success rate of 85%, completing consecutive three-fragment removal at 60% and at 95% with one human intervention. It further generalizes to unseen five-fragment scenarios with an average success rate of 80% for single-fragment removal. Project page: https://surgical-robotics.github.io/AGRO-SUVIDE/