Fully automated quantification of regional cerebral blood flow with three-dimensional stereotaxic region of interest template: Validation using magnetic resonance imaging - Technical note

The previously reported three-dimensional stereotaxic region of interest (ROI) template (3DSRT-t) for the analysis of anatomically standardized technetium-99m-L,L-ethyl cysteinate dimer (Tc-99m-ECD) single photon emission computed tomography (SPECT) images was modified for use in a fully automated regional cerebral blood flow (rCBF) quantification software, 3DSRT, incorporating an anatomical standardization engine transplanted from statistical parametric mapping 99 and ROIs for quantification based on 3DSRT-t. Three-dimensional T-2-weighted magnetic resonance images of 10 patients with localized infarcted areas were compared with the ROI contour of 3DSRT, and the positions of the central sulcus in the primary sensorimotor area were also estimated. All positions of the 20 lesions were in strict accordance with the ROI delineation of 3DSRT. The central sulcus was identified on at least one side of 210 paired ROIs and in the middle of 192 (91.4%) of these 210 paired ROIs among the 273 paired ROIs of the primary sensorimotor area. The central sulcus was recognized in the middle of more than 71.4% of the ROIs in which the central sulcus was identifiable in the respective 28 slices of the primary sensorimotor area. Fully automated accurate ROI delineation on anatomically standardized images is possible with 3DSRT, which enables objective quantification of rCBF and vascular reserve in only a few minutes using Tc-99m-ECD SPECT images obtained by the RVR method.