Membrane-peeling with a handheld tremor-cancelling robotic device for vitreoretinal surgery

Abstract

Purpose : Robot-assisted intraocular microsurgery aimed at dampening surgeon physiologic tremor may improve surgical manipulation of delicate micron-scale anatomical structures of the eye. We demonstrate the feasibility of Micron, a piezoelectric motor-operated handheld device, in facilitating microsurgical membrane peeling maneuvers via tremor cancellation.

Methods : The Micron 6-DoF with a bent-tip 20G needle as end-effector was used. An eye model consisting of a sorbothane rubber circle approximately 1 inch in diameter and 1/16th of an inch in height was used as the retinal model, with a 9 µm thick polyvinylidene chloride film demarcated in 9 locations to simulate the internal limiting membrane/epiretinal membrane to be peeled. Directional force was measured by a load cell which the rubber pad was mounted on. LED tracking of positional data as well as a calculated hard-stop were used to adjust the device for tremor cancellation. The average maximum upward and downward force with and without tremor cancellation and virtual fixture activation were recorded for 16 peeled films at a time, with equal distribution of trials aided by Micron. Paired and unpaired student t-tests were used for analysis using GraphPad Software.

Results : A total of 5 training sessions and 6 testing sessions were performed. The average maximum downward force with tremor cancellation was 21.9 mN in the training sessions and 12.3 in the testing sessions (p=0.0384). There was no statistically significant change in average upward force between the training and testing sessions, or between the trials with and without tremor cancellation in either group.

Conclusions : There was a statistically significant decrease in average maximum downward force with tremor cancellation between the training and testing sessions. Future studies in an animal model are warranted to better evaluate the feasibility of the device for intraocular surgery.

This is a 2020 ARVO Annual Meeting abstract.