Abstract
Background Augmented reality (AR) navigation has demonstrated accuracy for pedicle screw placement but has not been well studied for angular trajectories required to access disc spaces and sacroiliac joints (SIJs). This study evaluated the accuracy of AR-guided angular trajectories comparing 2 registration methods: intraoperative 3D computed tomography (CT)-based registration and 2D–3D registration using intraoperative radiographs matched to preoperative CT imaging.
Methods In this prospective cadaveric study, 4 experienced spine surgeons performed 67 AR-navigated angular trajectories targeting thoracic, lumbar, and SIJ anatomy across 4 human cadavers. Jamshidi tools were used to simulate angular navigation. Trajectories were planned using either intraoperative CT registration or 2D–3D registration. Absolute stereotactic accuracy was assessed by comparing postoperative CT trajectories to the system’s virtual trajectories. Angular deviations were measured in axial and sagittal planes with a predefined success criterion of a 99% upper confidence bound ≤3.0°.
Results All trajectories, including 32 medial, 31 lateral, and 4 intracanal (trajectories directed into the spinal canal to simulate access pathways for decompression or discal approaches), met predefined accuracy criteria for both methods. The 99% upper bounds of angular error were 1.65° (axial) and 1.54° (sagittal) for intraoperative registration and 1.78° (axial) and 1.75° (sagittal) for 2D–3D registration. Mean angular errors in both planes remained below 3.0°. Cadavers had a mean age of 72.5 years and a mean body mass index of 22.2.
Conclusion AR-guided navigation enables accurate angular trajectory guidance for disc space and SIJ targeting. Comparable accuracy between intraoperative CT and 2D–3D registration suggests that 2D–3D AR workflows may reduce radiation exposure and operative time while maintaining precision.
Clinical Relevance AR navigation may improve surgical workflow by providing real-time 3D visualization while reducing reliance on intraoperative CT, potentially decreasing radiation exposure and operative time without compromising accuracy in complex spine procedures.
Level of Evidence Level 5 (cadaveric study).
- augmented reality
- computer-navigated surgery
- neuronavigation
- cadaveric trial
- augmented reality-assisted surgery
- neuronavigation registration methods
Introduction
Precise navigation is critical in spine surgery to minimize tissue damage and enhance procedural safety.1,2 However, conventional navigation methods, including fluoroscopy and computer-assisted systems, present challenges such as attention shift, radiation exposure, and trajectory control limitations. These limitations are particularly relevant in interbody fusion procedures, such as transforaminal lumbar interbody fusion (TLIF), lateral lumbar interbody fusion (LLIF), extreme lateral interbody fusion, and sacroiliac joint (SIJ) fusions, which require precise, angular navigation to access targeted anatomical structures.3 These procedures require oblique, lateral, or combined entry angles, allowing surgeons to navigate around the lumbar plexus, thecal sac, and other critical structures to access the intervertebral disc space or SIJ.4 Some approaches require traversing bone, making trajectory control through denser structures essential.5
While traditional navigation improves patient safety and accuracy, its limitations of radiation exposure and line of sight interruptions warrant further investigation. Consistent execution of diverse approach angles is essential for surgical precision. Augmented reality (AR) technology has emerged as a promising alternative, offering real-time intraoperative 3-dimensional (3D) visualization to enhance accuracy.6,7 The Augmedics xvision AR navigation system has recently achieved US Food and Drug Administration approval for intradiscal navigation in addition to pedicle screw navigation.8 This computerized, image-guided navigation tool uses wireless optical tracking to display the real-time location of tracked surgical instruments relative to the patient’s scan directly in the surgical field.
The accuracy of AR guidance depends on how well the systems register to the patient’s anatomy. AR navigation typically relies on intraoperative 3D imaging registration from a computed tomography (CT) scan, but recent advancements allow for 2-dimensional (2D) intraoperative x-ray registration matched with preoperative CT scans. Although AR navigation has demonstrated high accuracy in pedicle screw placement, its application for interbody fusion and SIJ procedures, requiring diverse approach angles, remains underexplored.8 Trajectories in these procedures demand precise angulation, often navigating around critical neurovascular structures and traversing bone. Understanding the accuracy of AR-guided trajectories in these complex entry points is crucial for expanding its clinical utility.
The primary objective of this study was to evaluate the accuracy of AR-guided navigation under the external vision system (XVS) system using a Jamshidi needle model for accessing the disc space, spinal canal, and SIJ. Additionally, we compared trajectory accuracy using traditional 3D intraoperative registration vs the 2D–3D registration method.
Methods
Cadaveric Trial Design
The study was conducted in compliance with the ethical guidelines for human cadaver studies in Augmedics Headquarters (Chicago, IL). Four male human cadavers were procured. The full spine was utilized. All 4 cadavers had no history of spine surgery.9
The primary endpoint was to evaluate absolute stereotaxic navigation accuracy by comparing the final trajectory position on postoperative CT imaging with the virtual trajectory displayed by the XVS. Angular deviations of the positioned tool from the virtual trajectory were quantified in both the axial and sagittal planes. The acceptable upper limit of the 99% confidence interval (CI) of final trajectory position deviation from the virtual trajectory was set to be <3.0° in the axial and sagittal planes. Based on previous cadaver studies with a mean trajectory error of 1.41° ± 0.98° in the axial view and 1.44° ± 0.83° in the sagittal view (41 cervical screws), a minimum of 40 angular trajectories of navigated Jamshidi needles was set as an acceptable threshold.
Navigation and Registration Methods
Four senior spine surgeons (2 neurosurgeons and 2 orthopedic spine surgeons) performed all surgical and navigation procedures (Table 1). Each participating surgeon had extensive experience using spine navigation and was trained by Augmedics personnel on using the Xvision system and its 2 registration methods:
Intraoperative registration method: A validated approach that registers a 3D intraoperative CT scan to the patient’s anatomy.
2D–3D registration method (2D–3D): A method that matches intraoperative x-ray images with preoperative 3D CT scans to enable registration and spinal navigation. Alignment is achieved by analyzing geometric parameters such as image centering, distance, and distortion between a fixed C marker on the cadaver and a calibrated C-arm ring adapter during x-ray acquisition.
Surgeon’s information detailing their specialty and experience with navigation technology.
To reduce intersurgeon variability, each surgeon performed 13 to 22 angular navigation trajectories at different spinal levels using both registration methods on either the left or right side (Table 2 and Supplementary Figure 1). Distinct workflows were developed for each registration method (Figure 1).
Total number of trajectories by surgeon, spine level, and registration method.
Schematic of the surgical workflow demonstrating the 2 methods to register augmented reality navigation. The 2D–3D registration method was done on the left side of the cadaver, while intraoperative registration was done on the right. XVS, external vision system.
AR Surgical Navigation Technique
The test was performed in a room simulating an actual operating room (OR) under simulated use conditions. Each human cadaver was placed supine on an operating room table and then underwent CT preoperatively (Siemens SOMATOM CT scanner) with parameters set to bone setting as well as with 1 mm slicing and spacing. Angular trajectories were placed using the 2 XVS registration methods, that is, registration with intraoperative CT or registration of 2D intraoperative x-ray images with preoperative 3D CT images (2D–3D). For the 2D–3D registration method, at least 2 intraoperative x-ray images of the section to be instrumented and their correlating infrared (IR) images were acquired to then be matched with the preoperative 3D CT scan. For continuity, trajectories were placed on the right side of the cadaver using the intraoperative registration method and on the left side using the 2D–3D registration method. SI, lumbar, and thoracic trajectories were assessed when accessing the SIJ, spinal canal, and disc space (Figure 1).
Left-Sided 2D–3D Trajectories
For the SI and lumbar trajectories, a percutaneous pin (PP) and adapter were placed in the posterior superior iliac spine, and a patient marker was fixed to the PP adapter. Preoperative CT scans were loaded into the XVS, and automatic vertebral segmentation was performed. A 12″ C-arm with a fixated C-arm ring adapter was positioned above the SI and lumbar area, and at least 2 intraoperative x-ray images and their corresponding IR images were acquired. Registration was then completed between the 3D CT image and 2D x-ray images, creating a virtual 3D model of the segmented scanned area. The Jamshidi needles were navigated in angular trajectories, with 2 tools laterally placed in the SIJ and additional tools navigated in lateral and medial trajectories to the lumbar disc space. Each trajectory was saved, and a postoperative 3D intraoperative scan was performed for analysis.
For thoracic trajectories, a spinous process clamp was placed on L1 or L2, and a patient marker was fixed to it. A 12″ C-arm with a fixated C-arm ring adapter was positioned above the thoracic region, and at least 2 intraoperative x-ray images and their corresponding IR images were acquired. Registration between the 3D CT image and 2D x-ray images was completed, and a virtual 3D model of the segmented spine was displayed. Jamshidi needles were navigated into the disc space of 3 to 5 thoracic vertebrae, ensuring adequate spacing to prevent tool collisions. Once all needles were placed, a postoperative 3D intraoperative scan was performed for analysis.
Right-Sided Intraoperative Trajectories
For the SI and lumbar trajectories, a PP and adapter were placed in the posterior superior iliac spine, and the patient marker was secured to the PP adapter. An X-marker was positioned on the cadaver’s back, above the surgical area. A CT scan of the lumbosacral area was performed, followed by X-link registration, mapping the spatial position of the patient marker relative to the X-marker. A 3D model of the scanned area was created, and registration was completed. The Jamshidi needles were navigated into angular trajectories, with 2 tools laterally placed in the SIJ and 3 to 4 needles placed in lateral and medial lumbar disc trajectories. Once all trajectories were saved, postoperative 3D intraoperative CT was performed for analysis.
For thoracic trajectories, the patient marker was affixed to L1 or L2 via a spinous process clamp, and an X-marker was positioned above the thoracic surgical area. CT of the thoracic spine was performed, followed by X-link registration. A 3D model of the thoracic spine was created, and registration was finalized. Jamshidi tools were navigated into 8 thoracic vertebrae, with markers covered during placement to allow uninterrupted navigation. A postoperative 3D intraoperative scan was performed for analysis.
Accuracy Assessment and Statistical Analysis
The study included navigated trajectories over various angles in the lumbosacral and lower thoracic regions using the Xvision system. These trajectories were designed to evaluate accuracy in targeting the disc space and lateral trajectories to the SIJ. Some trajectories were performed using the intraoperative registration algorithm, while others were performed using the 2D–3D registration algorithm.
The absolute stereotactic navigation accuracy was measured by comparing the final trajectory position on postoperative CT with the virtual trajectory displayed by the XVS. To quantify accuracy, anatomical reference lines were drawn. Figure 2 illustrates the axial/sagittal angular-error measurement technique (β1 − β) using a representative cervical screw trajectory from a separate case; the identical measurement method was applied to all Jamshidi thoracic and lumbar disc-space and SIJ trajectories in this study. Angular deviations of the positioned trajectory from the virtual trajectory were measured in both the axial and sagittal planes, with an upper 99% CI threshold of 3.0° defined as acceptable.
Measurement of angular errors in the axial (A + C) and sagittal (B + D) planes. Comparison is made between intraoperative navigation images of virtual trajectory (A + B) and the final trajectory placement on post-op CT (C + D). Examples of recommended anatomical lines are drawn in white: axial plane (A + C)—mid-sagittal line (bisecting the vertebral body); sagittal plane (B + D)—along the superior endplate. Angular error is computed as (β1 − β). Figure 1 illustrates the axial/sagittal angular-error measurement technique (β1 − β) using a representative cervical screw trajectory from a separate case; the identical measurement method was applied to all Jamshidi thoracic and lumbar disc-space and sacroiliac joint trajectories in this study (see Methods).
For angular error calculations, the axial virtual trajectory angle was defined as the angle made by the virtual trajectory axis with the anatomical reference line (β, Figure 2a), while the axial postoperative trajectory angle was defined as the angle made by the positioned trajectory axis with the same reference line (β1, Figure 2c). The angular error in the axial plane was computed as the difference: (β1 − β). Similarly, the sagittal virtual trajectory angle was defined as the angle made by the virtual trajectory axis with the anatomical line of the respective vertebral body (β, Figure 2b), and the sagittal postoperative trajectory angle was defined as the angle made by the positioned trajectory axis with the anatomical line of the respective vertebral body (β1, Figure 2d). The angular error in the sagittal plane was computed as (β1 − β). All analyses were conducted in the plane of the tool using 3D Slicer software. Descriptive statistics were used to calculate the mean, SD, and 99% CIs for system accuracy.
Results
Cadaver Demographics and Surgeon Information
The tested cadavers were all men with a mean age of 72.5 years, an average height of 70 inches, an average body mass index (BMI) of 22.2, and no history of spine surgery (Table 3). The 4 attending spine surgeons conducting the trajectories had a total combined 52 years of surgical experience, ranging from 8 to 22 years. All had at least 1 year of experience using navigation systems, ranging from 1 to 15 years (Table 1).
Demographics of participating cadavers.
Accuracy Analysis
The number of trajectories broken down by surgeon, spine level, and registration method is listed in Table 2 (Supplementary Figure 1). A total of 32 medial trajectories and 31 lateral trajectories, as well as 4 intracanal trajectories, were performed, defined as trajectories directed into the spinal canal to simulate access pathways for decompression or discal approaches. There were 22 trajectories in the thoracic region, 29 in the lumbar region, and 16 in the SIJ. Moreover, there were 31 trajectories utilizing intraoperative registration and 36 trajectories utilizing 2D–3D registration.
Descriptive statistics were used to calculate the mean, SD, and 99% CIs of the system’s accuracy. The mean angular error for intraoperative registration was 1.23° (axial) and 1.17° (sagittal), with an SD of 0.90° (axial) and 0.80° (sagittal). The mean angular error for 2D–3D registration was 1.44° (axial) and 1.41° (sagittal), with an SD of 0.78° (axial) and 0.79° (sagittal; Table 4). Mean trajectory errors in both axial and sagittal planes for both registration methods were below the predefined threshold of 3.0° (Table 4). Furthermore, our data demonstrated that 15 trajectories fell below 1.0° and 25 trajectories fell below 2.0° of error in both axial and sagittal planes (Figure 3).
Angular trajectory accuracy analysis—angular errors in axial and sagittal views.
Accuracy target plot, demonstrating all 67 trajectories and their degree of deviation from both the sagittal and axial axes. All trajectories were below the predetermined 3.0° threshold but were further categorized to demonstrate additional accuracy details.
Discussion
We report the first evaluation of accuracy for the XVS in thoracic and lumbar disc space trajectories, as well as intracanal and SIJ trajectories. Of note, we were able to demonstrate 2 main points. First, regardless of registration method (intraoperative vs 2D−3D), AR with the head-mounted display can be accurately utilized to navigate multiple trajectories into the disc space for procedures such as TLIF, LLIF, and SIJ fusions. Second, the 2D–3D registration method was determined to be just as accurate as the intraoperative registration method, demonstrating the possibility for efficient use of x-rays with minimal radiation and reduction of surgical time when compared with the Intra-Op registration method, which needs a CT image to be obtained during surgery. Our results of 67 navigated trajectories revealed that the XVS achieved a high level of accuracy for both registration methods, with mean trajectory errors in both planes being below 3.0°.
Accuracy
These findings highlight the high level of accuracy achievable with this technology. Prior research on navigation-assisted spinal surgery has demonstrated trajectory accuracies as high as 94%, with mean angular differences of 2.17° ± 2.20° on axial images and 2.16° ± 2.24° on sagittal images in pedicle screw placement.10,11 Given these established benchmarks, our acceptance criteria were set to ≤3° to ensure comparable precision. Notably, we were able to demonstrate this level of accuracy across 4 different surgeons, 2 different registration methods, and 12 spinal levels on both sides of the body (Tables 1 and 2, Supplement 1). Future studies should focus on assessing the ability of AR to guide actual interbody placement and compare it with established navigation systems to further evaluate its clinical applicability.
Surgical Workflow
No literature exists on the use of AR to plan a trajectory for interbody requiring procedures like TLIF, with many currently using other 3D systems like StealthStation or Brainlab Spine Navigation to target the facet joints and disc space.12 When a procedure requires microscopic magnification and illumination (such as in tubular minimally invasive surgery [MIS]), the surgeon can switch to microscope-based AR displays, but this is currently only being described in case studies. Therefore, during microsurgery, to assess one’s location during surgery, the respective traditional navigation wand is frequently introduced into the surgical field, with the surgeon needing line-of-sight breakage to look at a 2D monitor for understanding of their location.13
Our data suggest that the high accuracy of interbody trajectory using AR might remove the need for other current navigation systems, as one could use this entire system for trajectory planning of screws, interbodies, and docking on facets or in the disc space. The surgeon can also further benefit from the 3D overlaid image during microscope use to identify relevant anatomy, which has been shown to be feasible, safe, and not dramatically additive to the literature’s report of surgical time needed to complete an MIS-TLIF, as in this case.14
By only using AR for both trajectory, 3D image overlay, and interbody placement guidance, one could further optimize surgical workflow by limiting the need to switch between various navigation systems. Future studies should assess this surgical workflow and its impact on surgical time, comparing it with current methods that use 3D systems other than AR for trajectory and rely solely on the microscope for dissection work, with frequent introduction of a 2D navigation system to identify relevant anatomy.
Reduced Radiation Exposure With the 2D–3D Registration Method
With the use of AR, the amount of radiation exposure to which surgeons and patients are subjected has been shown to be reduced when performing spinal surgeries.15 On average, 1.58 mSv is the amount of patient exposure during a MIS-TLIF.16 The 2D–3D registration method requires obtaining only 2 x-ray images during surgery and combines it with a preoperative CT scan, potentially further reducing radiation exposure during TLIF or SIJ fusion procedures.17 We found that this does not come at the cost of accuracy, as the 2D–3D registration was comparable to the intraoperative registration method, with both being accurate when a postoperative CT scan was subsequently compared with the intraoperative virtual tool. Unfortunately, in our study, we did not assess radiation exposure for each registration method and can only speculate on this benefit. Furthermore, the addition of AR navigation to the disc space and SIJ will reduce the need for extensive intraoperative radiography when targeting the disc space with traditional methods.
Application of AR for Angular Trajectories
Regarding AR’s use for guiding angular trajectories needed in performing disc space or SIJ work, we found it easier to remain focused during the entire case by not having to shift attention or line of sight away from the patient.9 All of the surgeons in this study had several years of experience using navigation systems for spine surgery, with only 1 having 1 year of experience prior to this experiment. On questioning, some points this surgeon described when asked about comparing AR with traditional 2D user interfaces included a reduction in the nonintuitive learning curve that comes with interpreting 2D views, mentioning that, with AR, they were able to directly visualize a 3D model, increasing their understanding of the trajectory at a faster rate. Future studies should assess this when it comes to new spine surgeons trialing AR for angular trajectory planning and performance in spine surgeries; they should also assess AR’s learning curve compared with traditional 2D navigation systems.
The accuracy demonstrated in this cadaveric study provides a strong foundation for proceeding to prospective clinical evaluation. Future clinical trials should assess AR-guided angular trajectories in MIS-TLIF, LLIF, and SIJ fusion cases, incorporating patient-reported outcomes, fusion rates, and complication data. The potential for a single AR platform to guide both trajectory planning and interbody placement could meaningfully streamline surgical workflows
Limitations
Several limitations should be considered when interpreting these findings. First, this study was conducted in a cadaveric model, which does not fully replicate in vivo surgical conditions, including soft tissue resistance, bleeding, and dynamic anatomical changes. As such, the ability to assess soft tissue traversal and real-time procedural constraints was limited, and clinical extrapolation should be made cautiously. Second, the sample size was relatively small and consisted of 4 cadavers with narrow demographic characteristics (older age and low body mass index), which may limit generalizability to broader patient populations, particularly those with higher BMI or varying anatomical complexity. Third, all participating surgeons were highly experienced with navigation systems, introducing potential performance bias. The accuracy observed in this study may not fully reflect outcomes among less experienced users or during the early learning curve of AR navigation adoption. Fourth, this study was conducted using a single AR navigation platform within a single-vendor setting, which may limit the generalizability of results to other navigation systems with different hardware, software, or registration workflows. Fifth, while the 2D–3D registration method demonstrated comparable accuracy to intraoperative CT-based registration, this study did not directly quantify operative time, workflow efficiency, or radiation exposure. Therefore, conclusions regarding reductions in radiation dose or procedural time remain inferential and should be validated in future prospective clinical studies. Finally, although this study focused on angular trajectory accuracy using Jamshidi needles as a surrogate for interventional access, it did not evaluate downstream procedural steps such as implant placement or clinical outcomes. Additionally, imaging documentation of needle trajectories and pathway validation was limited, as the angular error has a depth-dependent effect on absolute tip displacement: a given angular deviation produces proportionally greater positional error at deeper targets. Future studies should consider reporting both angular error and translational tip deviation, stratified by trajectory depth, to provide a more complete characterization of navigation accuracy for trajectories of varying lengths; therefore, future studies should incorporate more comprehensive visualization and verification of instrument pathways. Future prospective clinical studies should include patients with higher BMI, prior spinal surgery, and spinal deformity to enhance generalizability. Additionally, direct quantification of radiation exposure (eg, dose-area product) and operative time for each registration method would convert the current inferential conclusions into concrete data.
Conclusion
We report the first description of the Augmedics XVS being able to accurately display a trajectory for an interbody procedure like a TLIF, with an accuracy rate comparable to those of other computer navigation systems reported in the literature.18 These results highlight the potential of the Augmedics XVS to enhance precision in complex spine procedures, supporting its role in improving surgical outcomes through accurate 3D trajectory guidance and minimizing procedural risks.
Supplementary material
online supplementary file 1.
FIGURE S1.
Distribution of AR-guided trajectories across surgeons, spinal level, and registration method.Footnotes
Funding The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests The authors report no conflicts of interest in this work.
Patient Consent Donor specimens were obtained through the UC Davis Willed Body Program with documented informed consent for research use.
Ethics Considerations Cadaveric testing was performed under Augmedics internal anatomical-research protocols DHF-VAR-000018 and DHF-VAR-000020.
Data Availability Statement Reasonable requests for data can be sent to the corresponding author.
- This manuscript is generously published free of charge by ISASS, the International Society for the Advancement of Spine Surgery. Copyright © 2026 ISASS. To see more or order reprints or permissions, see http://ijssurgery.com.
References
- 1.↵Mummaneni PV , Haid RW , Rodts GE . Lumbar interbody fusion: state-of-the-art technical advances. J Neurosurg. 2004;1(1):24–30. 10.3171/spi.2004.1.1.0024
- 2.↵Ozgur BM , Aryan HE , Pimenta L , Taylor WR . Extreme lateral interbody fusion (XLIF): a novel surgical technique for anterior lumbar interbody fusion. Spine J. 2006;6(4):435–443. 10.1016/j.spinee.2005.08.012
- 3.↵Moore T , McLain RF . Image-guided surgery in resection of benign cervicothoracic spinal tumors: a report of two cases. Spine J. 2005;5(1):109–114. 10.1016/j.spinee.2004.06.020
- 4.↵Foley KT , Holly LT , Schwender JD . Minimally invasive lumbar fusion. Spine. 2003;28(15 Suppl):S26–S35. 10.1097/01.BRS.0000076895.52418.5E
- 5.↵Aranke M , McCrudy G , Rooney K , et al . Minimally invasive and conservative interventions for the treatment of sacroiliac joint pain: a review of recent literature. Orthop Rev. 2022;14(3):34098. 10.52965/001c.34098
- 6.↵Kim HJ , Yang JH , Chang D-G , et al . Adult spinal deformity: current concepts and decision-making strategies for management. Asian Spine J. 2020;14(6):886–897. 10.31616/asj.2020.0568
- 7.↵Amiot LP , Lang K , Putzier M , Zippel H , Labelle H . Comparative results between conventional and computer-assisted pedicle screw installation in the thoracic, lumbar, and sacral spine. Spine. 2000;25(5):606–614. 10.1097/00007632-200003010-00012
- 8.↵Bhatt FR , Orosz LD , Tewari A , et al . Augmented reality-assisted spine surgery: an early experience demonstrating safety and accuracy with 218 screws. Global Spine J. 2023;13(7):2047–2052. 10.1177/21925682211069321
- 9.↵Molina CA , Theodore N , Ahmed AK , et al . Augmented reality–assisted pedicle screw insertion: a cadaveric proof-of-concept study. J Neurosurg. 2019;31(1):139–146. 10.3171/2018.12.SPINE181142
- 10.↵Xi Z , Chou D , Mummaneni PV , Burch S . The navigated oblique lumbar interbody fusion: accuracy rate, effect on surgical time, and complications. Neurospine. 2020;17(1):260–267. 10.14245/ns.1938358.179
- 11.↵Miller CA , Ledonio CG , Hunt MA , Siddiq F , Polly DW . Reliability of the planned pedicle screw trajectory versus the actual pedicle screw trajectory using intra-operative 3D CT and image guidance. Int J Spine Surg. 2016;10:38. 10.14444/3038
- 12.↵Carl B , Bopp M , Saß B , et al . Reliable navigation registration in cranial and spine surgery based on intraoperative computed tomography. Neurosurg Focus. 2019;47(6):2019.8.FOCUS19621. 10.3171/2019.8.FOCUS19621
- 13.↵Faiella E , Castiello G , Bernetti C , et al . Impact of an augmented reality navigation system (SIRIO) on bone percutaneous procedures: a comparative analysis with standard CT-guided technique. Curr Oncol. 2021;28(3):1751–1760. 10.3390/curroncol28030163
- 14.↵Sommer F , Hussain I , Kirnaz S , et al . Augmented reality to improve surgical workflow in minimally invasive transforaminal lumbar interbody fusion—a feasibility study with case series. Neurospine. 2022;19(3):574–585. 10.14245/ns.2244134.067
- 15.↵
- 16.↵Kim CH , Lee CH , Kim KP . How high are radiation-related risks in minimally invasive transforaminal lumbar interbody fusion compared with traditional open surgery: a meta-analysis and dose estimates of ionizing radiation. Clin Spine Surg. 2016;29(2):52–59. 10.1097/BSD.0000000000000351
- 17.↵Dietrich TJ , Pfirrmann CWA , Schwab A , Pankalla K , Buck FM . Comparison of radiation dose, workflow, patient comfort and financial break-even of standard digital radiography and a novel biplanar low-dose x-ray system for upright full-length lower limb and whole spine radiography. Skeletal Radiol. 2013;42(7):959–967. 10.1007/s00256-013-1600-0
- 18.↵Torres J , James AR , Alimi M , Tsiouris AJ , Geannette C , Härtl R . Screw placement accuracy for minimally invasive transforaminal lumbar interbody fusion surgery: a study on 3-d neuronavigation-guided surgery. Global Spine J. 2012;2(3):143–152. 10.1055/s-0032-1326949









