Accuracy Evaluation Trial of Mixed Reality-Guided Spinal Puncture Technology

Abstract

Purpose: To evaluate the accuracy of mixed reality (MR)-guided visualization technology for spinal puncture (MRsp).

Methods: MRsp involved the following three steps: 1. Lumbar spine computed tomography (CT) data were obtained to reconstruct virtual 3D images, which were imported into a HoloLens (2nd gen). 2. The patented MR system quickly recognized the spatial orientation and superimposed the virtual image over the real spine in the HoloLens. 3. The operator performed the spinal puncture with structural information provided by the virtual image. A posture fixation cushion was used to keep the subjects' lateral decubitus position consistent. 12 subjects were recruited to verify the setup error and the registration error. The setup error was calculated using the first two CT scans and measuring the displacement of two location markers. The projection points of the upper edge of the L3 spinous process (L3↑), the lower edge of the L3 spinous process (L3↓), and the lower edge of the L4 spinous process (L4↓) in the virtual image were positioned and marked on the skin as the registration markers. A third CT scan was performed to determine the registration error by measuring the displacement between the three registration markers and the corresponding real spinous process edges.

Results: The setup errors in the position of the cranial location marker between CT scans along the left-right (LR), anterior-posterior (AP), and superior-inferior (SI) axes of the CT bed measured 0.09 ± 0.06 cm, 0.30 ± 0.28 cm, and 0.22 ± 0.12 cm, respectively, while those of the position of the caudal location marker measured 0.08 ± 0.06 cm, 0.29 ± 0.18 cm, and 0.18 ± 0.10 cm, respectively. The registration errors between the three registration markers and the subject's real L3↑, L3↓, and L4↓ were 0.11 ± 0.09 cm, 0.15 ± 0.13 cm, and 0.13 ± 0.10 cm, respectively, in the SI direction.

Conclusion: This MR-guided visualization technology for spinal puncture can accurately and quickly superimpose the reconstructed 3D CT images over a real human spine.

Keywords: immobilization; mixed reality; registration error; setup error; spinal puncture; superimposition.

Figure 1

HoloLens (2nd gen) hardware, Microsoft Corp., 2019.

Figure 2

In-house modified vacuum cushion for fixation in the lateral decubitus position.

Figure 3

MR-based virtual 3D models automatically adapted to one subject. Left: A subject in the lateral decubitus position with a QR code (a) and two beacons (b/c) affixed to the dorsal skin. Right: The position of the subject’s virtual 3D lumbar model was automatically and manually adapted to match the real lumbar spine. b’/c’: Virtual beacons. Yellow arrows: manual fine-tuning in four directions.

Figure 4

Accuracy trial workflow for MRsp.

Figure 5

Method of projecting the virtual image over the real spine. L3↑: Upper edge of the L3 spinous process. L3↓: Lower edge of the L3 spinous process. L4↓: Lower edge of the L4 spinous process.

Figure 6

Measurement of setup error on the three major anatomical planes on 3D CT. The projections of the cranial beacon in the first CT (purple circle) and the second CT (white circle) were compared, and the 3D displacements were calculated (a: transverse plane; b: coronal plane; c: sagittal plane).

Figure 7

Measurement of lumbar spine registration error on 3D CT. The displacement values between the real L3↑ (purple circle) and the virtual L3↑ registration marker (white circle) in the AP and SI directions of the bed were considered the registration error.