Recurrent multi-view 6DoF pose estimation for marker-less surgical tool tracking
International Journal of Computer Assisted Radiology and Surgery · 2025
DOI: 10.1007/s11548-025-03436-8 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Computer Assisted Radiology and Surgery, 2025. Verfasst von Niklas Agethen und 4 weiteren Autor:innen. Beteiligte Einrichtung: Fraunhofer MEVIS, Max-von-Laue-Str. 2, 28359, Bremen, Germany. niklas.agethen@mevis.fraunhofer.de. Hauptthemen laut MEDLINE: Deep Learning, Surgical Instruments, Surgery, Computer-Assisted. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 17 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Universität Bremen.
Abstract (Originalsprache)
Purpose
Marker-based tracking of surgical instruments facilitates surgical navigation systems with high precision, but requires time-consuming preparation and is prone to stains or occluded markers. Deep learning promises marker-less tracking based solely on RGB videos to address these challenges. In this paper, object pose estimation is applied to surgical instrument tracking using a novel deep learning architecture.
Methods
We combine pose estimation from multiple views with recurrent neural networks to better exploit temporal coherence for improved tracking. We also investigate the performance under conditions where the instrument is obscured. We enhance an existing pose (distribution) estimation pipeline by a spatio-temporal feature extractor that allows for feature incorporation along an entire sequence of frames.
Results
On a synthetic dataset we achieve a mean tip error below 1.0 mm and an angle error below 0.2 $$^{\circ }$$ ∘ using a four-camera setup. On a real dataset with four cameras we achieve an error below 3.0 mm. Under limited instrument visibility our recurrent approach can predict the tip position approximately 3 mm more precisely than the non-recurrent approach.
Conclusion
Our findings on a synthetic dataset of surgical instruments demonstrate that deep-learning-based tracking using multiple cameras simultaneously can be competitive with marker-based systems. Additionally, the temporal information obtained through the architecture’s recurrent nature is advantageous when the instrument is occluded. The synthesis of multi-view and recurrence has thus been shown to enhance the reliability and usability of high-precision surgical pose estimation.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Fraunhofer MEVIS, Max-von-Laue-Str. 2, 28359, Bremen, Germany. niklas.agethen@mevis.fraunhofer.de.
- University of Bremen, Bibliothekstraße 1, 28359, Bremen, Germany.
- Fraunhofer MEVIS, Max-von-Laue-Str. 2, 28359, Bremen, Germany.
Förderung
- Universität Bremen
Zitieren
APA
Agethen, N., Rosskamp, J., Koller, T. L., Klein, J., & Zachmann, G. (2025). Recurrent multi-view 6DoF pose estimation for marker-less surgical tool tracking. International Journal of Computer Assisted Radiology and Surgery. https://doi.org/10.1007/s11548-025-03436-8
Vancouver
Agethen N, Rosskamp J, Koller TL, Klein J, Zachmann G. Recurrent multi-view 6DoF pose estimation for marker-less surgical tool tracking. International Journal of Computer Assisted Radiology and Surgery. 2025. doi:10.1007/s11548-025-03436-8
BibTeX
@article{Agethen2025,
author = {Agethen, Niklas and Rosskamp, Janis and Koller, Tom L and Klein, Jan and Zachmann, Gabriel},
title = {Recurrent multi-view 6DoF pose estimation for marker-less surgical tool tracking},
journal = {International Journal of Computer Assisted Radiology and Surgery},
year = {2025},
doi = {10.1007/s11548-025-03436-8}
}
EUVIMED