Surgical Navigation System for Pedicle Screw Placement Based on Mixed Reality

Inaccurate screw insertion has been a major concern in pedicle screw placement in spinal surgery because of the potential for postoperative complications. Efforts to improve screw insertion have led to the development of novel surgical navigation systems. However, current surgical navigation systems...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:International journal of control, automation, and systems automation, and systems, 2023-12, Vol.21 (12), p.3983-3993
Hauptverfasser: Hwang, Seokbin, Lee, Suk-joong, Kim, Sungmin
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Inaccurate screw insertion has been a major concern in pedicle screw placement in spinal surgery because of the potential for postoperative complications. Efforts to improve screw insertion have led to the development of novel surgical navigation systems. However, current surgical navigation systems have several problems in that the attention of surgeons is frequently interrupted during surgery, and complications occur in their depth perception because of the limitations of 2D monitors. This study proposes a surgical navigation system to address these issues using mixed reality (MR)-based smart glasses. We developed a navigation system that provides 3D visualization of the surgical tool with respect to the patient’s anatomy. It utilizes preoperative surgical planning data to provide visual feedback and guidance. An orthopedic surgeon performed pedicle screw placement with the navigation system in two different environments, using Microsoft HoloLens (HoloLens) and a 2D monitor. The results were evaluated according to procedural time, translational error, angular error, and clinical accuracy. The mean procedural time was 111.3 ± 52.7 s with the HoloLens and 192.1 ± 104.0 s with the 2D monitor. The mean translational error was 2.14 ± 1.13 mm at the entry and 3.14 ± 0.90 mm at the target with HoloLens. With the 2D monitor, the mean translational error was 2.10 ± 0.97 mm and 3.41 ± 2.16 mm at the entry and the target, respectively. The mean angular error was 6.44 ± 1.94 deg with HoloLens and 7.14 ± 4.20 deg with the 2D monitor. All screws were inserted intrapedicularly in both environments. The navigation system enables free visualization, reflects the human eye’s perspective, and retains the advantages of MR-based smart glasses. A navigation system compatible with minimally invasive surgery should be developed in the future.
ISSN:1598-6446
2005-4092
DOI:10.1007/s12555-023-0083-6