Visualizing diffusion tensor images of the mouse spinal cord

Abstract

Within biological systems water molecules undergo continuous stochastic Brownian motion. The rate of this diffusion can give clues to the structure of underlying tissues. In some tissues the rate is anisotropic -faster in some directions than others. Diffusionrate images are second-order tensor fields and can be calculated from diffusion-weighted magnetic resonance images. A 2D diffusion tensor image (DTI) and an associated anatomical scalar field, created during the tensor calculation, define seven values at each spatial location. Visually representing these images is a challenge because they contain so many inter-related components. We present two new methods for visually representing DTIs. The first method displays an array of ellipsoids where the shape of each ellipsoid represents one tensor value. The novel aspect of this representation is that the ellipsoids are all normalized to approximately the same size so that they can be displayed simultaneously in context. The second method uses concepts from oil painting to represent the seven-valued data with multiple layers of varying brush strokes. Both methods successfully display most or all of the information in DTIs and provide exploratory methods for understanding them. The ellipsoid method has a simpler interpretation and explanation than the painting-motivated method; the painting-motivated method displays more of the information and is easier to read quantitatively. We demonstrate the methods on images of the mouse spinal cord. The visualizations show significant differences between spinal cords from mice suffering from Experimental Allergic Encephalomyelitis (EAE) and spinal cords from wild-type mice. The differences are consistent with differences shown histologically and suggest that our new non-invasive imaging methodology and visualization of the results could have early diagnostic value for neurodegenerative diseases.

Biomedical Visualization

“…Alternatively, the visualization can be restricted to a single slice. For this type of representation Laidlaw et al proposed a method to normalize the size of the ellipsoids and another method that leverages techniques from oil painting to offer a complete view of the tissue anisotropy on a given slice [91].…”

Dense Glyph Sampling for Visualization

“…The question of placing these glyphs has been subject of discussion in several contexts. The most common strategies are regular sampling, random sampling with or without Poisson property [17,13] or procedural texture generation, e.g., using reaction diffusion. In vector field visualization, Turk and Banks proposed a method to place arrows along streamlines generated by streamline optimization [23].…”

“…The use of glyphs, ranging from simple ellipses to more advanced glyphs as superquadrics [10], is commonly done for visualizing such data sets. The glyphs are mostly placed in grid points or are randomly spread [17]. Figure 8(b) shows a result using our sample generation.…”

An Introduction to Visualization of Diffusion Tensor Imaging and Its Applications

“…Laidlaw et al normalized the size of the ellipsoids to fit more of them in a single image [32] (see figure 8(a)). While this method forgoes the ability to show mean diffusivity, it creates more uniform glyphs that show anatomy and pathology over regions better than the non-normalized ellipsoids.…”

“…While this method forgoes the ability to show mean diffusivity, it creates more uniform glyphs that show anatomy and pathology over regions better than the non-normalized ellipsoids. Laidlaw et al [32] also developed a method that uses the concepts of brush strokes and layering from oil painting to emphasize the diffusion patterns. They used 2D brush strokes both individually, to encode specific values, and collectively, to show spatial connections and to generate texture and a sense of speed corresponding to the speed of diffusion.…”