Frameless Patient Tracking With Adhesive Optical Skin Markers for Augmented Reality Surgical Navigation in Spine Surgery.

STUDY DESIGN: Observational study. OBJECTIVE: The aim of this study was to evaluate the accuracy of a new frameless reference marker system for patient tracking by analyzing the effect of vertebral position within the surgical field. SUMMARY OF BACKGROUND DATA: Most modern navigation systems for spine surgery rely on a dynamic reference frame attached to a vertebra for tracking the patient. This solution has the drawback of being bulky and obstructing the surgical field, while requiring that the dynamic reference frame is moved between vertebras to maintain accuracy. METHODS: An augmented reality surgical navigation (ARSN) system with intraoperative cone beam computed tomography (CBCT) capability was installed in a hybrid operating room. The ARSN system used input from four video cameras for tracking adhesive skin markers placed around the surgical field. The frameless reference marker system was evaluated first in four human cadavers, and then in 20 patients undergoing navigated spine surgery. In each CBCT, the impact of vertebral position in the surgical field on technical accuracy was analyzed. The technical accuracy of the inserted pedicle devices was determined by measuring the distance between the planned position and the placed pedicle device, at the bone entry point. RESULTS: The overall mean technical accuracy was 1.65 ±â€Š1.24 mm at the bone entry point (n = 366). There was no statistically significant difference in technical accuracy between levels within CBCTs (P ≥ 0.12 for all comparisons). Linear regressions showed that null- to negligible parts of the effect on technical accuracy could be explained by the number of absolute levels away from the index vertebrae (r ≤ 0.007 for all, ß ≤ 0.071 for all). CONCLUSION: The frameless reference marker system based on adhesive skin markers is unobtrusive and affords the ARSN system a high accuracy throughout the navigated surgical field, independent of vertebral position. LEVEL OF EVIDENCE: 3.

Machine learning for automated 3-dimensional segmentation of the spine and suggested placement of pedicle screws based on intraoperative cone-beam computer tomography.

OBJECTIVE: The goal of this study was to develop and validate a system for automatic segmentation of the spine, pedicle identification, and screw path suggestion for use with an intraoperative 3D surgical navigation system. METHODS: Cone-beam CT (CBCT) images of the spines of 21 cadavers were obtained. An automated model-based approach was used for segmentation. Using machine learning methodology, the algorithm was trained and validated on the image data sets. For measuring accuracy, surface area errors of the automatic segmentation were compared to the manually outlined reference surface on CBCT. To further test both technical and clinical accuracy, the algorithm was applied to a set of 20 clinical cases. The authors evaluated the system's accuracy in pedicle identification by measuring the distance between the user-defined midpoint of each pedicle and the automatically segmented midpoint. Finally, 2 independent surgeons performed a qualitative evaluation of the segmentation to judge whether it was adequate to guide surgical navigation and whether it would have resulted in a clinically acceptable pedicle screw placement. RESULTS: The clinically relevant pedicle identification and automatic pedicle screw planning accuracy was 86.1%. By excluding patients with severe spinal deformities (i.e., Cobb angle > 75° and severe spinal degeneration) and previous surgeries, a success rate of 95.4% was achieved. The mean time (± SD) for automatic segmentation and screw planning in 5 vertebrae was 11 ± 4 seconds. CONCLUSIONS: The technology investigated has the potential to aid surgeons in navigational planning and improve surgical navigation workflow while maintaining patient safety.