Comparing Fiducial-Based and Intraoperative Computed Tomography-Based Registration for Frameless Stereotactic Brain Biopsy.

OBJECTIVE: The aim of this work was to compare fiducial-based and intraoperative computed tomography (iCT)-based registration for frameless stereotactic brain biopsy. METHODS: Of 50 frameless stereotactic biopsies with the VarioGuide, 30 cases were registered as iCT based and 20 as fiducial based. Statistical analysis of the target registration error (TRE), dose length product, effective radiation dose (ED), operation time, and diagnostic yield was performed. RESULTS: The mean TRE was significantly lower using iCT-based registration (mean ± SD: 0.70 ± 0.32 vs. 2.43 ± 0.73 mm, p < 0.0001). The ED was significantly lower when using iCT-based registration compared to standard navigational CT (mean ± SD: 0.10 ± 0.13 vs. 2.23 ± 0.34 mSv, p < 0.0001). Post-biopsy iCT was associated with a significant lower (p < 0.0001) ED compared to standard CT (mean ± SD: 1.04 ± 0.18 vs. 1.65 ± 0.26 mSv). The mean surgical time was shorter using iCT-based registration, although the mean total operating room (OR) time did not differ significantly. The diagnostic yield was 96.7% (iCT group) versus 95% (fiducial group). Post-biopsy imaging revealed severe bleeding in 3.3% (iCT group) versus 5% (fiducial group). CONCLUSION: iCT-based registration for frameless stereotactic biopsies increases the accuracy significantly without negative effects on the surgical time or the overall time in the OR. Appropriate scan protocols in iCT registration contribute to a significant reduction of the radiation exposure. The high accuracy of the iCT makes it the more favorable registration strategy when taking biopsies of small tumors or lesions near eloquent brain areas.

Indocyanine Green Angiography Visualized by Augmented Reality in Aneurysm Surgery.

OBJECTIVE: We prospectively investigated how to integrate indocyanine green (ICG) angiography in an augmented reality (AR) setting for aneurysm surgery. METHODS: In 20 patients with a total of 22 aneurysms, the head-up display of the operating microscope (Kinevo900) was used for AR. ICG-AR was established directly by the head-up display superimposing the ICG angiography as green live video overlay. In addition, the reconstructed outline of the three-dimensional (3D) vessel architecture was visualized by AR applying intraoperative low-dose computed tomography (vessel-AR). RESULTS: In all patients, ICG-AR and vessel-AR were successfully implemented. The flow in the vessels could be observed directly in the white light view of the microscope oculars without being distracted from the surgical site by looking on separate screens. This factor enabled also surgical manipulation during ICG angiography. In parallel, AR additionally visualized the 3D vessel architecture, enhancing the understanding of the 3D anatomy (target registration error, 0.71 ± 0.21 mm; intraoperative low-dose computed tomography effective dose, 42.7 µSv). Linear (n = 28; range, 1-8.5 mm) and rotational (n = 3; range, 2.9°-14.4°) navigation adjustments performed in 18 of 20 patients resulted in a close matching of the vessel-AR outline with the real vessel situation after preparation, compensating for shifting. CONCLUSIONS: ICG-AR could be successfully implemented. It facilitated surgical manipulation and flow interpretation during ICG angiography because it could be observed directly while looking through the microscope oculars in white light instead of being distracted from the surgical site while looking on separate screens. Additional AR visualizing the vessel architecture improved understanding of 3D anatomy for preparation and clipping.

Microscope-Based Augmented Reality in Degenerative Spine Surgery: Initial Experience.

OBJECTIVE: To establish microscope-based augmented reality (AR) support for degenerative spine surgery. METHODS: Head-up displays of operating microscopes were used to establish AR in a series of 10 patients. Segmentation of the vertebra and additional target structures, which were visualized by AR, was based on preoperative magnetic resonance and computed tomography (CT) images, that were nonrigidly fused to low-dose intraoperative CT (iCT) data. AR registration was achieved by automatic registration applying iCT and microscope calibration. RESULTS: AR support could be smoothly implemented in the surgical workflow. AR allowed to visualize the target structures reliably in the surgical field, facilitating surgical orientation. Flexible placement of the reference array enabled AR implementation for anterior, lateral, posterior median, and posterior paramedian approaches. Identification of bony and artificial landmarks allowed validating registration accuracy; the measured target registration error was 1.11 ± 0.42 mm (mean ± standard deviation). The effective dose for registration scanning ranged from 0.52 to 8.71 mSv, which is on average about one-third of a standard diagnostic spine scan. This depended mainly on the scan length (mean scan length cervical/thoracic/lumbar: 99/218/118 mm). Longest scan ranges were in the mid-thoracic region to ensure unambiguous vertebra assignment as prerequisite for reliable nonlinear registration (mean cervical/thoracic/lumbar effective dose: 0.52/6.14/2.99 mSv). CONCLUSIONS: Reliable microscope-based AR support is possible because of automatic registration based on intraoperative imaging. Application of AR in degenerative spine surgery has a big potential; it might be especially helpful in complex anatomical situations and resident education.

Navigated 3-Dimensional Intraoperative Ultrasound for Spine Surgery.

OBJECTIVE: To integrate 3-dimensional (3D) intraoperative ultrasound (iUS) data in spinal navigation. METHODS: In 11 patients with intradural spinal tumors, 3D-iUS was performed before and after tumor resection. Intraoperative computed tomography (iCT) was used for automatic patient registration for spinal navigation; fiducial-based registration was performed in 1 case. The outlines of the vertebra were defined in preoperative image data by automatic mapping; risk and target structures were segmented manually; all these data were rigidly and if necessary non-rigidly registered with iCT. For 3D-iUS acquisition, tracked convex-shaped transducers (contact surface: 29 x 10 mm; scanning frequency: 10-3.8 MHz or 13-5 MHz) were used. RESULTS: Navigated 3D-iUS was successfully implemented in all cases; 3D-iUS datasets were acquired and could be used as 3D image data for further navigation after iUS scanning. The 3D objects defined in preoperative image data, outlining the vertebra, target and risk structures, could be visualized in the 3D-iUS data. Navigated 3D-iUS allowed to reliably evaluate the extent of resection in all cases and updating of navigation, ensuring high navigational accuracy. The target registration error applying iCT-based automatic registration was 0.78 ± 0.23 mm. The effective dose for iCT was 0.11 ± 0.077 mSv for cervical and 1.75 ± 0.72 mSv for thoracic scans. CONCLUSIONS: Using 3D-iUS can be successfully integrated in spinal navigation. Automatic registration applying low-dose iCT and non-linear image registration offers displaying preoperative images in the same orientation as the 3D-iUS scan, as well as visualizing segmented structures in the navigated 3D-iUS data. This greatly facilitates image interpretation. Navigated 3D-iUS provides a possibility for navigation updating and immediate online quality control.