Virtual Environments in Surgery: Synthetic SLAM Validation in Knee Arthroplasty - 25.24
Title:
Virtual Environments in Surgery: Synthetic SLAM Validation in Knee Arthroplasty
Authors:
Arne SCHIERBAUM 1, Tobias NEISS-THEUERKAUFF 2, Thomas LUHMANN 1, Frank WALLHOFF 2, Till SIEBERTH 1
1 Institute for Applied Photogrammetry and Geoinformatics, Jade University of Applied Sciences, Oldenburg, Germany;
2 Institute for Technical Assistive Systems, Jade University of Applied Sciences, Oldenburg, Germany
Keywords:
3d scanning, SLAM, Blender, simulation, surgery, knee arthroplasty
Abstract:
Knee arthroplasty is a commonly performed surgical procedure in which computer-assisted and partially robot-guided systems are increasingly used to improve precision. Traditionally, these systems rely on optical markers that are fixed to the femur and tibia. However, these invasive markers require drilling, which can prolong the healing process and increase the risk of infection. This work aims to lay the foundation for markerless navigation, eliminating the need for such fixation.
To achieve this, the visible surface of the knee is captured using SLAM (Simultaneous Localization and
Mapping) with a handheld trinocular camera system. Challenges include the low-texture surface, reflections caused by wet surfaces, and the movement of the knee during surgery. Evaluating the accuracy of the SLAM-based systems is difficult, due to too few suitable test datasets and the limited availability of real 3D medical data. In addition, realistic annotated images of bones are missing, which are necessary for AI-based masking the knee during SLAM.
This paper presents a simulation environment developed using Blender, in which surgical scenes are created based on anatomical 3D models. The system simulates camera motion and generates image data for knee reconstruction, which can be evaluated against known ground truth. The simulation not only supports the geometric optimization of the camera system but also provides direct access to the image position of the bone. As a result, it eliminates the need for separate bone segmentation, which in real scenarios is typically performed using deep learning methods that remain prone to error. These segmentation inaccuracies can significantly impact SLAM performance and make its evaluation more difficult. By generating precise masks alongside the synthetic images, the simulation avoids this source of uncertainty and enables a more accurate and isolated assessment of SLAM algorithms. At the same time, the simulation environment is used to automatically generate training data for segmentation and to improve masking.
Abstract:
Presentation:
VIDEO will be available here in Q3.2026.
VIDEO availble in proceedings (purchase order)
How to Cite (MLA):
A. Schierbaum et al., "Virtual Environments in Surgery: Synthetic SLAM Validation in Knee Arthroplasty", Proceedings of 3DBODY.TECH 2025 - 16th International Conference and Expo on 3D/4D Body Scanning, Data and Processing Technologies, Lugano, Switzerland, 21-22 Oct. 2025, #24
Details:
Proceedings: 3DBODY.TECH 2025, 21-22 Oct. 2025, Lugano, Switzerland
Paper/Presentation: #24
DOI: -
License/Copyright notice
Proceedings: © Hometrica Consulting - Dr. Nicola D'Apuzzo, Switzerland, hometrica.ch.
Authors retain all rights to individual papers, which are licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
The papers appearing in the proceedings reflect the author's opinions. Their inclusion in the proceedings does not necessary constitute endorsement by the editor or by the publisher.
Note: click the + on the top left of the page to open/close the menu.