Research Article: MRI, CT and high resolution macro-anatomical images with cryosectioning of a Beagle brain: Creating the base of a multimodal imaging atlas

Date Published: March 7, 2019

Publisher: Public Library of Science

Author(s): Kálmán Czeibert, Gábor Baksa, András Grimm, Szilvia Anett Nagy, Enikő Kubinyi, Örs Petneházy, Andreas Hahn.


Most common methods that directly show macro- or microscopic anatomy of the brain usually require the removal of the organ from the neurocranium. However, the brain can be revealed in situ by using proper sectioning techniques. Our aim was to both improve the cryosectioning method, test its limits and create a high-resolution macro-anatomical image series of a Beagle brain, which at the time of the study did not exist. A two-year-old female Beagle has been scanned with CT and MRI ante and post mortem, then the arteries of the head were filled with red resin. After freezing to -80°C, a neurocranium block was created and was embedded into a water-gelatin mix. Using a special milling device and a DSLR camera, 1112 consecutive RGB-color cryosections were made with a 100 μm layer thickness and captured in high resolution (300 dpi, 24-bit color, and pixel size was 19.5 x 19.5 μm). Image post-processing was done with Adobe Photoshop CS3 and Thermo Scientific Amira 6.0 softwares, and as a result of the proper alignment and coregistration, visualization and comparing was possible with all the applied imaging modalities (CT, MRI, cryosectioning) in any arbitrary plane. Surface models from the arteries, veins, brain and skull were also generated after segmentation in the same coordinate system, giving a unique opportunity for comparing the two-dimensional and three-dimensional anatomy. This is the first study which focuses directly to this high-definition multimodal visualization of the canine brain, and it provides the most accurate results compared to previous cryosectioning studies, as using an improved method, higher image quality, more detailed image, proper color fidelity and lower artefact formation were achieved. Based on the methodology we described, it can serve as a base for future multimodal (CT, MR, augmented- or virtual reality) imaging atlases for medical, educational and scientific purposes.

Partial Text

Creating a novel anatomical atlas requires a technique showing the organ in a different perspective (e.g. detailedness, staining, and tissue-fidelity), or that the applied method results in enhanced image quality compared to a previous atlas. In order to validate the need for cryosectioning of a canine brain to create a new comparative image series, we provide an overview of the main techniques that make possible considering the requirements mentioned above. There are several ways to visualize macro- or microanatomical structures in anatomy: conventional preparations and sections can be made shortly post mortem on a fresh cadaver, or previously fixed with a fixative agent [1], creating macerated bones and skeletons [2–4], or corrosion casting [5–7]. The result of the tissue preparation can be captured in photographs or videos, or these procedures can be combined with the different imaging methods (like CT or MRI), so that the selected region or specimen can be digitized for 2- and 3-dimensional analysis. These techniques are all suitable methods to visualize the different systems of the body, and they can be grouped by direct/indirect methods and tissue maintaining/tissue destructive methods. Direct imaging means that the original tissue can be seen either in its real color (e.g. with endoscopy) or dyed using histological staining [8]. Indirect imaging, such as computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET) and single-photon emission computed tomography (SPECT) provides a computer-generated picture (most commonly a grayscale image), and during post-processing an artificial color is added for differentiation based on tissue properties. Examples include X-ray attenuation (CT), or spin and precessing of the proton and water diffusion (MR) [9]. Indirect imaging methods are tissue maintainers, but have some limitations to show the real environment (mostly due to small spatial resolution). The central nervous system is enclosed into a bony capsule, and in the case of larger animals and humans the head cannot be sectioned together with the skull due to the hardness of the bone—although the present methods make it possible to produce histological sections as large as a complete human brain [10,11]. Plastination of the slices is a good method for tissue preservation and selective staining is possible [12,13], but the specimen’s color is partly altered during the procedure and shrinkage could occur [14–16].

Using high resolution photography, the brain and the surrounding structures were visualized with high acuity and detailedness, as it is shown in examples zooming on the ethmoturbinates, intrinsic lingual muscles and structures of the middle cranial fossa (Figs 3 and 4).

One of the main advantages of cryosectioning is that any region of the body can be studied, regardless of the tissue composition including thick bones that can be easily cut. Therefore, high resolution RGB-images can be obtained, and if the layer thickness is small enough, multiplanar reconstructions (MPR) can be made in any arbitrary plane by using specially designed softwares (e.g. Amira or 3DSlicer). When structural imaging techniques (MR, CT) are associated, the volumes can be registered together, thus multimodal atlases can be easily generated in the same coordinate system. As ante mortem functional studies, tractography of neural pathways (Diffusion Tensor Imaging, DTI), PET (Positron-Emission Tomography) and SPECT (Single Photon Emission Computed Tomography) imaging could also be done prior cryosectioning. The segmentation of individual structures or systems makes it possible to create three-dimensional models beneficial to several fields, including education (e.g. understanding the relationship between the 2D images series and the 3D-structure), research (comparative neuroanatomy) and clinical work (reference atlas for studying the conservative brains regions during surgical planning). As in the current study the brain was not removed from the neurocranial cavity, which is usually required to allow histological or macroscopic slicing [42,45], the cryosectioned structures can be seen in their original position (depending on the occurrence of freezing expansion or shrinkage). For example, the vessels around the brain and the nerves, which are entering and leaving the skull through different foramina can be traced around the osseous structures and the central nervous system. As no fixative agent was used in our study, and only the arteries were filled with red polyurethane (without producing extravasates), all the tissues can be seen in their original color. This was also confirmed by using a color checker passport, which helps with camera and image calibration during the post-processing procedure of the cryosectioned images.

The improved method utilized for cryomacrotomisation in the current study has proved to be successful to serve as a reliable base for a comparative, multimodal brain imaging atlas. As the cryosectioning procedure is not equivalent with a histological study, but it represents a macro-anatomical cross-sectional approach, the desirable resolution for future studies depends on the aim of the study: we believe that the resolution we used is enough for reliable comparison with the CT and MRI series, and for the segmentation of major anatomical structures. When investigating smaller structures and thinner layers (below 20–30 micrometres) we recommend the standard histological procedures. This is the first study in dogs that has visualized the brain using comparative imaging modalities resulting in excellent quality and detailedness (high resolution images with 0.1 mm layer thickness). Possibilities also include the application of 3D-modeling and 3D-printing to enhance learning in graduate and postgraduate studies. In the results section we showed the main advantages of the improved cryosectioning technique, and gave several examples through multiplanar reconstructions how it can be an aid in the comparative imaging by merging the diagnostic imaging modalities (CT, MR) with cryosectioned images and three-dimensional models. We also plan to use cryo-protective agent before freezing the block in future works. Due to the detailedness of the images, the image sets are suitable for selective structure extraction and exact segmentation, as was shown with the skull, brain, arteries and the veins. Thus, the resulting models can be exported, zoomed into and studied in any perspective, which would be an excellent support to accompany textbooks on the subject matter. We are also considering the fact of inter-subject variability would be good to complete these studies results, thus obtaining cryosectioned image series from dogs with dolichocephalic, mesocephalic and brachycephalic skull types are advantageous. Finally, the cryosectioning technique used, provides a unique tool for examining any parts of the body, no matter the hardness of the tissues. As a result, future works can use this technique in order to provide helpful material for educational, scientific and medical purposes.




Leave a Reply

Your email address will not be published.