EUVIMEDEUVIMED entdecken
Crossref

Markerless 3D ultrasound reconstruction for augmented reality assisted minimally invasive spine surgery

Ruixuan Li, Yuyu Cai, Ayoob Davoodi, Gianni Borghesan, Carlos Rodriguez-Guerrero, Emmanuel Vander Poorten

International Journal of Computer Assisted Radiology and Surgery · 17.05.2026 · 21 (6) · 1301-1310

DOI: 10.1007/s11548-026-03684-2 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Computer Assisted Radiology and Surgery, online erschienen am 17.05.2026. Verfasst von Ruixuan Li und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Mechanical Engineering, Robot-Assisted Surgery Group, KU Leuven, Leuven, Belgium. ruixuan.li@kuleuven.be. Hauptthemen laut MEDLINE: Imaging, Three-Dimensional, Augmented Reality, Minimally Invasive Surgical Procedures, Surgery, Computer-Assisted. Die Arbeit nennt 21 Literaturangaben. Gefördert von: Fonds Wetenschappelijk Onderzoek.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

PURPOSE: Minimally invasive spine surgery (MISS) requires 3D navigation of bony anatomy, yet conventional navigation is limited by radiation or line-of-sight-dependent external trackers. This study developed a markerless augmented reality ultrasound (AR-US) system to provide radiation-free 3D visualization while eliminating tracker occlusion. METHODS: An integrated pipeline was developed to track a handheld US probe using a head-mounted display (HMD). The probe pose was estimated using FoundationPose during scanning. To mitigate inherent noise in the raw estimates, an Extended Kalman Filter was implemented for refinement. Spatially tracked 2D US images were subsequently used to reconstruct a 3D volume, which was rendered on the HMD. RESULTS: The proposed system reconstructed 3D volumes with an average representation error of 2.17 ± 1.57 mm on different experimental phantoms. The 3D visualization was also successfully rendered on the HMD, qualitatively demonstrating a spatial alignment with key anatomic features on a spine phantom. CONCLUSION: This work demonstrates the feasibility of a markerless AR-US system for coarse navigation in MISS. The presented approach offers a streamlined and intuitive solution, providing clinicians with 3D anatomy for navigating instrumentation..

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Imaging, Three-DimensionalAugmented RealityMinimally Invasive Surgical ProceduresSurgery, Computer-AssistedSpineHumansPhantoms, ImagingUltrasonography

Stichwörter der Autor:innen

3D reconstructionAugmented realityMinimally invasive surgeryUltrasound navigation

Autor:innen und Institutionen

  • Department of Mechanical Engineering, Robot-Assisted Surgery Group, KU Leuven, Leuven, Belgium. ruixuan.li@kuleuven.be.
  • Department of Mechanical Engineering, Robot-Assisted Surgery Group, KU Leuven, Leuven, Belgium.
  • Flanders Make @KU Leuven, Leuven, Belgium.

Förderung

  • Fonds Wetenschappelijk Onderzoek

Zitieren

APA

Li, R., Cai, Y., Davoodi, A., Borghesan, G., Rodriguez-Guerrero, C., & Vander Poorten, E. (2026). Markerless 3D ultrasound reconstruction for augmented reality assisted minimally invasive spine surgery. International Journal of Computer Assisted Radiology and Surgery, 21(6), 1301-1310. https://doi.org/10.1007/s11548-026-03684-2

Vancouver

Li R, Cai Y, Davoodi A, Borghesan G, Rodriguez-Guerrero C, Vander Poorten E. Markerless 3D ultrasound reconstruction for augmented reality assisted minimally invasive spine surgery. International Journal of Computer Assisted Radiology and Surgery. 2026;21(6):1301-1310. doi:10.1007/s11548-026-03684-2

BibTeX
@article{Li2026,
  author = {Li, Ruixuan and Cai, Yuyu and Davoodi, Ayoob and Borghesan, Gianni and Rodriguez-Guerrero, Carlos and Vander Poorten, Emmanuel},
  title = {Markerless 3D ultrasound reconstruction for augmented reality assisted minimally invasive spine surgery},
  journal = {International Journal of Computer Assisted Radiology and Surgery},
  year = {2026},
  volume = {21},
  number = {6},
  pages = {1301--1310},
  doi = {10.1007/s11548-026-03684-2}
}
Kostenlos starten →