Affordable Access

Publisher Website

Trocar localisation for robot-assisted vitreoretinal surgery.

Authors
  • Birch, Jeremy1
  • Da Cruz, Lyndon2
  • Rhode, Kawal3
  • Bergeles, Christos3
  • 1 School of Biomedical Engineering and Imaging Sciences, King's College London, Strand, London, WC2R 2LS, UK. [email protected].
  • 2 Moorfields Eye Hospital, 162 City Rd, London, EC1V 2PD, UK.
  • 3 School of Biomedical Engineering and Imaging Sciences, King's College London, Strand, London, WC2R 2LS, UK.
Type
Published Article
Journal
International Journal of Computer Assisted Radiology and Surgery
Publisher
Springer-Verlag
Publication Date
Feb 01, 2024
Volume
19
Issue
2
Pages
191–198
Identifiers
DOI: 10.1007/s11548-023-02987-y
PMID: 37354219
Source
Medline
Keywords
Language
English
License
Unknown

Abstract

Robot-assisted vitreoretinal surgery provides precise and consistent operations on the back of the eye. To perform this safely, knowledge of the surgical instrument's remote centre of motion (RCM) and the location of the insertion point into the eye (trocar) is required. This enables the robot to align both positions to pivot the instrument about the trocar, thus preventing any damaging lateral forces from being exerted. Building on a system developed in previous work, this study presents a trocar localisation method that uses a micro-camera mounted on a vitreoretinal surgical forceps, to track two ArUco markers attached on either side of a trocar. The trocar position is the estimated midpoint between the markers. Experimental evaluation of the trocar localisation was conducted. Results showed an RMSE of 1.82 mm for the localisation of the markers and an RMSE of 1.24 mm for the trocar localisation. The proposed camera-based trocar localisation presents reasonable consistency and accuracy and shows improved results compared to other current methods. Optimum accuracy for this application would necessitate a 1.4 mm absolute error margin, which corresponds to the trocar's radius. The trocar localisation results are successfully found within this margin, yet the marker localisation would require further refinement to ensure consistency of localisation within the error margin. Further work will refine these position estimates and ensure the error stays consistently within this boundary. © 2023. The Author(s).

Report this publication

Statistics

Seen <100 times