EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Benchmarking complete-to-partial point cloud registration techniques for laparoscopic surgery

Alberto Neri, Veronica Penza, Nazim Haouchine, Leonardo S Mattos

Frontiers in Robotics and AI · 17.11.2025 · 12

DOI: 10.3389/frobt.2025.1702360 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Robotics and AI, online erschienen am 17.11.2025. Verfasst von Alberto Neri und 3 weiteren Autor:innen. Beteiligte Einrichtung: Biomedical Robotics Lab, Advanced Robotics, Istituto Italiano di Tecnologia, Genoa, Italy. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIBIB NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

Registering a preoperative 3D model of an organ with its actual anatomy viewed from an intraoperative video is a fundamental challenge in computer-assisted surgery, especially for surgical augmented reality. To address this, we present a benchmark of state-of-the-art deep learning point-cloud registration methods, offering a transparent evaluation of their generalizability to surgical scenarios and establishing a robust guideline for developing advanced non-rigid algorithms.

Methods

We systematically evaluate traditional and deep learning GMM-based, correspondence-based, correspondence-free, matching-based, and liver-specific point cloud registration approaches on two surgical datasets: a deformed IRCAD liver set and DePoll dataset. We also propose our complete-to-partial point cloud registration framework that leverages keypoint extraction, overlap estimation, and a Transformer-based architecture, culminating in competitive registration results.

Results

Experimental evaluations on deformed IRCAD tests reveal that most deep learning methods achieve good registration performances with TRE<10 mm, MAE(R) < 4 and MAE(t)<5 mm. On DePoll, however, performance drops dramatically due to the large deformations.

Conclusion

In conclusion, deep-learning rigid registration methods remain reliable under small deformations and varying partiality but lose accuracy when faced with severe non-rigid changes. To overcome this, future work should focus on building non-rigid registration architectures that preserve the strengths of self-, cross-attention and overlap modules while enhancing correspondence estimation to handle large deformations in laparoscopic surgery.

Schlagwörter

Stichwörter der Autor:innen

computer-assisted surgerycorrespondencesdeep learninglaparoscopypoint cloud registration

Autor:innen und Institutionen

  • Biomedical Robotics Lab, Advanced Robotics, Istituto Italiano di Tecnologia, Genoa, Italy.
  • Department of Computer Science, Bioengineering, Robotics and Systems Engineering (DIBRIS), University of Genoa, Genova, Italy.
  • Harvard Medical School and Brigham and Women's Hospital, Boston, MA, United States.

Förderung

  • NIBIB NIH HHS

Zitieren

APA

Neri, A., Penza, V., Haouchine, N., & Mattos, L. S. (2025). Benchmarking complete-to-partial point cloud registration techniques for laparoscopic surgery. Frontiers in Robotics and AI, 12. https://doi.org/10.3389/frobt.2025.1702360

Vancouver

Neri A, Penza V, Haouchine N, Mattos LS. Benchmarking complete-to-partial point cloud registration techniques for laparoscopic surgery. Frontiers in Robotics and AI. 2025;12. doi:10.3389/frobt.2025.1702360

BibTeX
@article{Neri2025,
  author = {Neri, Alberto and Penza, Veronica and Haouchine, Nazim and Mattos, Leonardo S},
  title = {Benchmarking complete-to-partial point cloud registration techniques for laparoscopic surgery},
  journal = {Frontiers in Robotics and AI},
  year = {2025},
  volume = {12},
  doi = {10.3389/frobt.2025.1702360}
}
Kostenlos starten →