Synchrotron radiation-based micro computed tomography dataset - in vivo mouse brain - CA019 - peri-mortem

Contrast-enhanced imaging of peri-mortem morphological changes at CLS BMIT beamline This datasets features a synchrotron radiation-based micro computed tomography dataset of a C57BL/6J mouse injected with barium-based contrast agent into the lateral ventricle. Data was recorded at the beamline BMIT of the Canadian Light Source in November 2023. For more information, see the associated publication, repository or the FABRIC4 portal. subjCA019_scan005_pump_off_alive_vertical_001_scan_000.zipFull 3D-stack, time point 1, 58 min after start of contrast agent infusion subjCA019_scan005_pump_off_alive_vertical_001_scan_000_segmentation.zipFull 3D-stack, segmentation of the cerebral ventricles in a registered version of the above dataset.0 = background, 255 = ventricles subjCA019_scan004_pump_off_alive_vertical_001_scan_000_ROIaqueduct.zip3D-stack of a registered small region of interest containing the cerebral aqueduct subjCA019_scan005_pump_off_dying_vertical_025_scan_000.zipFull 3D-stack, time point 25, 79 min after start of contrast agent infusion, 4 min after pentobarbital euthanasia subjCA019_scan005_pump_off_dying_vertical_025_scan_000_segmentation.zipFull 3D-stack, segmentation of the cerebral ventricles in a registered version of the above dataset.0 = background, 255 = ventricles subjCA019_scan005_pump_off_dying_vertical_025_scan_000_ROIaqueduct.zip3D-stack of a registered small region of interest containing the cerebral aqueduct subjCA019_ROIaqueduct_differenceimage.zipDifference image, where the images of subjCA019_scan005_pump_off_dying_vertical_025_scan_000_ROIaqueduct.zip were subtracted from the images of subjCA019_scan004_pump_off_alive_vertical_001_scan_000_ROIaqueduct.zip Methods C57BL/6J (stock 000664) mice were supplied by Jackson Labs. To ensure proper acclimatization, the animals were housed within the local animal facility one week before the experiments. The experimental protocols were reviewed and approved by the responsible ethics committee of the University of Saskatchewan, approval number AUP 20230072 MOD#1. For the dataset shown, a male mouse (SubjectID: CA019) twelve weeks of age and 28.5 g body weight was first injected subcutaneously with buprenorphine (0.1 mg/kg) for analgesia. Anesthesia was induced after onset of analgesia 30 min later, via intraperitoneal injection of a cocktail of ketamine (50 mg/kg) and medetomidine (1 mg/kg). The animals were kept warm using heating pads. The depth of anesthesia was monitored by testing reflexes, and additional intraperitoneal injections were given as needed. Eye ointment was applied and the skull, neck, and upper thoracic region of the mouse were shaved to avoid potential artefacts during X-ray imaging. 2 × 0.5 ml glucose 10% was administered in two separate subcutaneous injections, and tracheotomy was performed using a custom, radiotransparent, 3D-printed plastic tracheal cannula with Y-adapter. The mouse was then secured in a stereotactic frame and ventilated with a MiniVent, providing 96% oxygen from an oxygen accumulator device and 0.5 – 2% isoflurane. The exhaled air was routed through a water column to maintain 2 cmH2O positive end-expiratory pressure. Stroke volume was set to 150 µl, respiration rate to 140 bpm.For cannula implantation, excess periosteum from the bone was removed and the bregma was identified. A small hole of about 1 mm diameter was drilled through the parietal bone at the injection coordinates. An MRI-compatible PEEK infusion cannula of 2.3 mm length was filled with standard concentration contrast agent 320 mg Ba/ml and implanted into the right lateral ventricle. Cannula coordinates were 1.2 mm lateral and 0.03 mm caudal of the bregma. The animal was then transferred to the radiation hutch containing the SRµCT imaging setup, where it was imaged at a photon energy of 37.8 keV, provided by the beamline’s double bent Laue crystal monochromator. 2 000 radiographs over a rotation range of 180° were acquired with a pco.edge 4.2 camera coupled with a tandem lens system (105 mm f/2.4 and 85 mm f/1.4) and a 200 µm LuAg:Ce scintillator for 8 µm effective pixel size. The field of view was reduced to 2048 × 1200 pixels due to the limited size of the X-ray beam, and recorded with 15 ms exposure time and 62 frames per second, meaning 1 ms overhead time. Acquisition time per scan was 32 s. Sample-detector distance was 0.6 m, source-sample distance was 57.8 m. The in vivo scan was started 58 min after start of the contrast agent infusion. Contrast agent flow rate was ramped up linearly from 0 to 0.25 µl/min within the first minute and kept until 5 µl of contrast agent were infused. To avoid backflow of CSF into the infusion cannula, a very low flow rate of 0.02 µl/min was maintained thereafter, meaning an overall volume of 5.75 µl had been infused at the time of acquiring the in vivo image. The infusion was then stopped 60 min after infusion start, and the animal was euthanized with an intraperitoneal overdose of sodium pentobarbital (300 µl, 240 mg/ml) 74 min after start of infusion. Time of death was as determined at 78 min as defined by peak end-tidal CO2 dropping to 0.1% as the most reproducible measurement. The post-mortem image was acquired 79 min after infusion start, or 1 min after death. Tomograms were reconstructed using the image processing toolkit tofu, which employs the UFO framework as a back-end. Center-of-rotation was determined automatically. Output image intensity values correspond to µΔx, the product of the linear attenuation coefficients µ and the voxel length Δx. In order to isolate potential motion of the ventricles from the effect of whole animal motion, data volumes of the different time points used for comparison were registered to the first reconstruction of the series (called reference image) via rigid registration. An extended bone mask without ventricular spaces was created for the reference image to ensure that the contrast enhancement of the ventricles would not affect registration. The registration was driven by maximizing normalized cross-correlation as an image similarity measure within the extended bone mask. Automatic image registration used the open-source software elastix (version 4.9) Images were rescaled to the intensity range in the ROI of the reference image to reduce quantification errors when using 256 bins for covering the intensity range during registration. Images of the original intensities were afterwards transformed based on the registration result. To calculate ventricular volumes in live and post-mortem states, the masks Mlive and Mpost-mortem were generated by automatic thresholding of the non-denoised and median-filtered images, respectively. For Mpost-mortem, a median filter with a 1.5-pixel radius spherical kernel was applied prior to thresholding. In both cases, thresholding was performed using Otsu’s method. Bones were excluded by applying the mask Mbones using a logical AND NOT operation. Connected component analysis (6-voxel connectivity) with manual component selection was then performed, followed by morphological closing with a 7.5-pixel-radius spherical kernel to fill in the unenhanced ChP located within the ventricles. Volume estimates for the ventricular masks were calculated using the material statistics module in Amira 3D, based on the number of voxels multiplied by the voxel volume. Cross-sectional areas in each coronal slice were obtained using the area module, which calculates the number of non-zero pixels multiplied by the pixel area.