Accuracy and Precision of a Surgical Navigation System: Effect of Camera and Patient Tracker Position and Number of Active Markers

• Published in: The open orthopaedics journal 2017-05, Vol.11 (1), p.493-501
• Main Authors: Gundle, Kenneth R, White, Jedediah K, Conrad, Ernest U, Ching, Randal P
• Format: Article
• Language: English
• ISSN: 1874-3250; 1874-3250
• DOI: 10.2174/1874325001711010493

Surgical navigation systems are increasingly used to aid resection and reconstruction of osseous malignancies. In the process of implementing image-based surgical navigation systems, there are numerous opportunities for error that may impact surgical outcome. This study aimed to examine modifiable sources of error in an idealized scenario, when using a bidirectional infrared surgical navigation system. Accuracy and precision were assessed using a computerized-numerical-controlled (CNC) machined grid with known distances between indentations while varying: 1) the distance from the grid to the navigation camera (range 150 to 247cm), 2) the distance from the grid to the patient tracker device (range 20 to 40cm), and 3) whether the minimum or maximum number of bidirectional infrared markers were actively functioning. For each scenario, distances between grid points were measured at 10-mm increments between 10 and 120mm, with twelve measurements made at each distance. The accuracy outcome was the root mean square (RMS) error between the navigation system distance and the actual grid distance. To assess precision, four indentations were recorded six times for each scenario while also varying the angle of the navigation system pointer. The outcome for precision testing was the standard deviation of the distance between each measured point to the mean three-dimensional coordinate of the six points for each cluster. Univariate and multiple linear regression revealed that as the distance from the navigation camera to the grid increased, the RMS error increased (p