Intraoperative image updating for brain shift following dural opening

Abstract

OBJECTIVE Preoperative magnetic resonance images (pMR) are typically coregistered to provide intraoperative navigation, the accuracy of which can be significantly compromised by brain deformation. In this study, the authors generated updated MR images (uMR) in the operating room (OR) to compensate for brain shift due to dural opening, and evaluated the accuracy and computational efficiency of the process. METHODS In 20 open cranial neurosurgical cases, a pair of intraoperative stereovision (iSV) images was acquired after dural opening to reconstruct a 3D profile of the exposed cortical surface. The iSV surface was registered with pMR to detect cortical displacements that were assimilated by a biomechanical model to estimate whole-brain nonrigid deformation and produce uMR in the OR. The uMR views were displayed on a commercial navigation system and compared side by side with the corresponding coregistered pMR. A tracked stylus was used to acquire coordinate locations of features on the cortical surface that served as independent positions for calculating target registration errors (TREs) for the coregistered uMR and pMR image volumes. RESULTS The uMR views were visually more accurate and well aligned with the iSV surface in terms of both geometry and texture compared with pMR where misalignment was evident. The average misfit between model estimates and measured displacements was 1.80 ± 0.35 mm, compared with the average initial misfit of 7.10 ± 2.78 mm between iSV and pMR, and the average TRE was 1.60 ± 0.43 mm across the 20 patients in the uMR image volume, compared with 7.31 ± 2.82 mm on average in the pMR cases. The iSV also proved to be accurate with an average error of 1.20 ± 0.37 mm. The overall computational time required to generate the uMR views was 7-8 minutes. CONCLUSIONS This study compensated for brain deformation caused by intraoperative dural opening using computational model-based assimilation of iSV cortical surface displacements. The uMR proved to be more accurate in terms of model-data misfit and TRE in the 20 patient cases evaluated relative to pMR. The computational time was acceptable (7-8 minutes) and the process caused minimal interruption of surgical workflow.

Keywords: FEM model; FRE = fiducial registration error; GPU = graphics processing unit; OR = operating room; RMS = root mean square; TRE = target registration error; brain deformation; diagnostic and operative techniques; iMR = intraoperative magnetic resonance imaging scanner; iSV = intraoperative stereovision; iUS = intraoperative ultrasound; image-guided neurosurgery; intraoperative stereovision; pMR = preoperative magnetic resonance images; sparse data; uMR = updated magnetic resonance images.

FIG. 1

Flowchart of the model-based image updating process. FEM = finite element method.

FIG. 2

Illustration of surface registration between iSV and pMR for Cases 8, 10, 15, and 16 (columns left to right). The first and second rows show the iSV texture maps and pMR-encoded images, respectively. Thresholded binary iSV (green) and pMR (red) images are overlaid in the third row, and misalignment indicates lateral shift of the cortical surface. The fourth row presents an overlay of iSV (green) and pMR (red) images after registration, in which the 2D displacements appear as white vectors. The fifth row shows an overlay of the iSV (colored) and pMR (grayscale) brain surfaces where the extracted 3D displacements appear as blue vectors (pointing from pMR to iSV). Figure is available in color online only.

FIG. 3

Intraoperative screenshot of the comparison views on the StealthStation from Case 16. Upper and lower rows show pMR and uMR, respectively. The green crosshairs represent the corresponding position in MR image space, which correctly points to the brain surface in uMR, but falls on the scalp in pMR. The green contours show boundaries of tumor, which was segmented prior to the start of surgery based on pMR. The large discrepancy between the contour lines and the contrast-enhanced region in uMR indicates that due to brain deformation the contours were no longer accurate after dural opening. Figure is available in color online only.

FIG. 4

TRE was evaluated by comparing tracked locations of dominant features on the cortical surface against their corresponding positions in uMR. Three feature points (in A–C) were touched with a tracked stylus probe, and manually identified in the uMR-encoded texture map (crosses, D). Figure is available in color online only.

FIG. 5

Comparison of pMR and uMR in 2D and 3D views from Cases 8, 10, 15, and 16 (columns left to right). The first and second rows show representative pMR and uMR slices overlaid with iSV (yellow contour lines), respectively, whereas the third and fourth rows present 3D views of the same overlays. Arrows in the 3D views highlight features that align accurately in uMR but poorly in pMR. Figure is available in color online only.

FIG. 6

Contour map of displacement magnitude in the direction normal to the cortical surface from Case 8. Contour lines show outward displacements with magnitudes of 1–9 mm. Figure is available in color online only.