EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Occlusion-Resilient Markerless Registration Using Probabilistic Depth Confidence Modeling in AR-Assisted Surgery

Michael A. Johnson, Daniel K. Thompson, Emily R. Carter

INTERNATIONAL JOURNAL OF RESEARCH & SOCIAL SCIENCES · 01.05.2026 · 6 (1) · 164-171

DOI: 10.66238/ijrss.v6i1.102 ↗

Kurz gefasst

Fachartikel in der Zeitschrift INTERNATIONAL JOURNAL OF RESEARCH & SOCIAL SCIENCES, online erschienen am 01.05.2026. Verfasst von Michael A. Johnson und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Intraoperative occlusion caused by surgical instruments often degrades depth-based markerless registration. This study introduces a probabilistic depth confidence model that dynamically weights valid surface regions during alignment. A pixel-level confidence map is estimated using temporal consistency and surface normal stability metrics. Weighted ICP optimization is then performed to prioritize reliable depth regions. Experiments were conducted on 32 simulated surgical sequences with varying occlusion ratios (10%–45%). Under 40% occlusion, registration error increased only by 0.42 mm compared with occlusion-free scenarios, whereas conventional ICP exhibited a 1.63 mm degradation. The framework maintained stable tracking at 40 FPS, demonstrating strong robustness to instrument interference.

Zitieren

APA

Johnson, M. A., Thompson, D. K., & Carter, E. R. (2026). Occlusion-Resilient Markerless Registration Using Probabilistic Depth Confidence Modeling in AR-Assisted Surgery. INTERNATIONAL JOURNAL OF RESEARCH & SOCIAL SCIENCES, 6(1), 164-171. https://doi.org/10.66238/ijrss.v6i1.102

Vancouver

Johnson MA, Thompson DK, Carter ER. Occlusion-Resilient Markerless Registration Using Probabilistic Depth Confidence Modeling in AR-Assisted Surgery. INTERNATIONAL JOURNAL OF RESEARCH & SOCIAL SCIENCES. 2026;6(1):164-171. doi:10.66238/ijrss.v6i1.102

BibTeX
@article{Johnson2026,
  author = {Johnson, Michael A. and Thompson, Daniel K. and Carter, Emily R.},
  title = {Occlusion-Resilient Markerless Registration Using Probabilistic Depth Confidence Modeling in AR-Assisted Surgery},
  journal = {INTERNATIONAL JOURNAL OF RESEARCH & SOCIAL SCIENCES},
  year = {2026},
  volume = {6},
  number = {1},
  pages = {164--171},
  doi = {10.66238/ijrss.v6i1.102}
}
Kostenlos starten →