Brain volumetry: A quantitative approach for detecting region‐specific structural abnormalities

Abstract

Dear Editor, In recent years, the conspicuous development and spread of imaging techniques, i.e. magnetic resonance imaging (MRI) and computed tomography (CT), have allowed non-invasive investigation of the three-dimensional (3D) gross anatomical structures of the brain. Volumetry is a quantitative approach using imaging techniques for detecting structural differences in organs among two or more groups. In our recent study, T2-weighted MRI-based volumetry revealed a dose-dependent volume reduction of the whole cerebellum by prenatal exposure to a single whole body X-irradiation in rats (Sawada et al. 2013). Here, we attempted the CT-based volumetry of the brain of a growth-retarded mouse (grt/ grt), which is known as an animal model for primary congenital hypothyroidism (Sasaki et al. 2007).Brains from male homozygous (grt/grt, n = 5) and heterogeneous (grt/+, n = 5) mice at 15 weeks of age were soaked in 150 mg/mL nonionic iodinated contrast agent (Iopamiron, Bayer Schering Pharma, Japan) in a 7.5% paraformaldehyde solution at 4°C for 14 days to achieve a better contrast of the white and gray matter structures in inverted images by ex vivo micro-CT (Saito and Murase 2012). Then, CT images were acquired using 3D micro-CT (RmCT, Rigaku, Tokyo, Japan) with a resolution of 39 × 39 × 39 μm 3 , a tube voltage peak of 90 kV, and a tube current of 200 μA with a scan time of 8 min. Brain regions, i.e. the cerebrum, diencephalon, midbrain, pons, medulla oblongata and cerebellum, were semi-automatically segmented on CT images using the "Morpho" tool of SliceOmatic software ver 4.3 (TomoVision, Montreal, Canada) based on image contrast as well as the user's knowledge of the anatomy. 3D-rendered images of the brain were constructed using those segmented images. Furthermore, segmented areas of each brain region were measured, and the volumes were calculated by multiplying the combined areas by the slice thickness (39 μm). The total volume of segmented regions was regarded as the whole-brain volume.Computed tomography images could distinguish gray and white matter structures by their contrast, and the brain regions were grossly identifiable (Fig. 1A). 3D-rendered images revealed no obvious difference in the 3D morphology and structural organization of the brain between grt/grt and grt/ + (Fig. 1B). The volume of whole-brain was calculated based on CT images. A significantly decreased brain volume was noted in grt/grt (P < 0.05) (Fig. 1C). Such brain hypoplasia was attributed to a region-specific lower volume of the grt/grt cerebrum by two-way ANOVA and post-hoc testing (P < 0.001) (Fig. 1D).While grt/grt mice have a defect in tyrosylprotein sulfotransferase 2 (Tpst2) gene, the cerebral volume reduction revealed in the present volumetric analysis may be rather involved in the hypothyroid status of grt/grt mice (Kobayashi et al. 2001), because similar cerebral hypoplasia is obtained by experimentallyinduced hypothyroid rats (Madeira et al. 1990). Thus, CT-or MRIbased volumetry can be screened changes in brain structures of smal...

Regional hypoplasia of somatosensory cortex in growth‐retarded mice (grt/grt)

“…CT‐based volumetry was conducted according to our previous procedures (Horiuchi‐Hirose et al ; Sawada et al ). Cerebral cortices were semi‐automatically segmented on CT images using the “Morpho” tool of SliceOmatic software ver 4.3 (TomoVision, Montreal, Canada), based on image contrast as well as the user's knowledge of anatomy.…”