SimPS-Net: Simultaneous Pose & Segmentation Network of Surgical Tools

4 pages•Published: March 8, 2024

Abstract

The ability to detect and localise surgical tools using RGB cameras during robotic assisted surgery can allow for the development of various implementations, such as vision- based active constraints and refinements in robot path planning, which can ultimately lead in improved patient safety during operation. For this purpose, the proposed network, SimPS-Net capable of both detection and 3D pose estimation of standard surgical tools using a single RGB camera, is introduced. In addition to the network, a novel dataset generated for training and testing is presented. The proposed network achieved a mean DICE coefficient of 85.0%, while also exhibiting a low average error of 5.5mm and 3.3◦ for 3D position and orientation respectively, thus outperforming the competing networks.

Keyphrases: 3D pose estimation, image segmentation, monocular, Surgical Tool Detection, surgical tool localisation

In: Joshua W Giles (editor). Proceedings of The 22nd Annual Meeting of the International Society for Computer Assisted Orthopaedic Surgery, vol 6, pages 90--93

Links:	https://easychair.org/publications/paper/Hznz
	https://doi.org/10.29007/plzv