Viewing modes

Single slice is a cross-sectional view of the XRH image stack along a selected plane.

XRH datasets are normally oriented (resliced) in a way that a scroll through the stack along the XY plane emulates the physical histology slicing of the tissue. This way, any XY single slice through the image stack (blue plane in the image below) is parallel to the histology cassette (when applicable) and the XY slice-scroll runs from the wax-block's surface towards the cassette (similarly to the knife on physical sectioning).

Orthogonal to the XY plane are the XZ and YZ planes (red and green planes in the image). XY, XZ and YZ are also referred to as orthogonal planes or ortho-planes.

μCT slices are single channel images, which usually represents intensity value of the captured signal. Although μCT slices are greyscale, false colouring can be applied through look up tables.

Dimensions

• Single voxel: Depending on the acquisition, reconstruction and/or processing method, voxels in μCT images can be isotropic (x = y = z) or anisotropic (x ≠ y ≠ z). Unless otherwise stated, XRH voxels should be considered isotropic.

• Single slice: In a X x Y x Z volume, each slice is represented by X x Y voxels. For a voxel size = n (μm), due to the isotropicity of the voxels each slice's dimensions are: Xxn x Yxn x 1xn

Thick slice is a 2D viewing mode that allows rendering of the voxel intensity of multiple consecutive slices of a slice-stack onto a single slice, based on specific criteria or operations. A thick slice image represents the result of these criteria or operations.

A thick slice can be generated by applying any of the following projection methods across a pre-selected number of slices (n), which defines the slice thickness.

• Maximum Intensity Projection (MIP) - MIP analyses each x, y voxel in the image across n slices (x,y,1 , x,y,2 , ... , x,y,n) along the z-axis and renders the voxel with the maximum value onto the viewing plane.

• Average Intensity Projection (Avg) - Avg analyses each x, y voxel in the image across n slices (x,y,1 , x,y,2 , ... , x,y,n) along the z-axis and rendered the average value of those voxels onto the viewing plane.

• Sum - Sum analyses each x, y voxel in the image across n slices (x,y,1 , x,y,2 , ... , x,y,n) along the z-axis and rendered the sum of the value of those voxels onto the viewing plane.

Thick slice mode is also referred to as Z-projection (Fiji) or Slab view (Dragonfly)

Thick slice roll is a 2D thick slice rendering that allows rolling of a pre-selected number of slices (n) along the z-axis of the stack. A single thick slice roll is accomplished by adding +1 slice to the first and n-th element of the first thick slice and reapplying the criteria or operations to the new thick slice.

Thick slice roll can be saved as a video file or image stack for interactive viewing of the data.

Example:

Assume an image stack containing 500 slices, and a user-defined thick slice thickness of 20 slices. The criterion the user has selected is MIP and the user is after a thick slice roll. In this case:

• the first thick slice stack image will be a MIP projection of slices 1-20

• the second one a MIP projection of slices 2-21,

• ..., and

• the last image will be a MIP of 481-500

consequently the 20x thick slice image stack will contain: 500-20 =480 thick slices.

Volumetric rendering is 3D visualisation mode used for displaying the grid of voxels contained in the XRH dataset.

Direct volume rendering algorithms visualise images of volumetric datasets by mapping data values to opacity and colour, without explicitly extracting geometric surfaces of the object.

Each voxel corresponds to a location in data space and has one or more data values associated with it. In the case of a CT reconstruction volumetric rendering visualises the 3D map of appropriately scaled linear attenuation coefficients.

Similarly to the 2D Thick slice, MIP rotation is generated by an algorithm that casts rays of light through the dataset, where only the voxel with the highest X-ray attenuation (i.e. the brightest voxel) along each ray path is rendered.

In this mode, the dataset is rotated about a user-defined axis, and the rotation allows for depth and perspective interpretation of the brightest features.