Till Riffert

Plausibility Tracking: A method to evaluate anatomical connectivity and microstructural properties along fiber pathways

Diffusion MRI is a non-invasive method that potentially gives insight into the brains white matter structure regarding the pathway of connections and properties of the axons. Here, we propose a novel global tractography method named Plausibility Tracking that provides the most plausible pathway, modeled as a smooth spline curve, between two locations in the brain. Compared to other tractography methods, plausibility tracking combines the more complete connectivity pattern of probabilistic tractography with smooth tracks that are globally optimized using the fiber orientation density function and hence is relatively robust against local noise and error propagation. It has been tested on phantom and biological data and compared to other methods of tractography. Plausibility tracking provides reliable local directions all along the fiber pathways which makes it especially interesting for tract-based analysis in combination with direction dependent indices of diffusion MRI. In order to demonstrate this potential of plausibility tracking, we propose a framework for the assessment and comparison of diffusion derived tissue properties. This framework comprises atlas-guided parameterization of tract representation and advanced bundle-specific indices describing fiber density, fiber spread and white matter complexity. We explore the new method using real data and show that it allows for a more specific interpretation of the white matters microstructure compared to rotationally invariant indices derived from the diffusion tensor.

Beyond fractional anisotropy: Extraction of bundle-specific structural metrics from crossing fiber models

Diffusion MRI (dMRI) measurements are used for inferring the microstructural properties of white matter and to reconstruct fiber pathways. Very often voxels contain complex fiber configurations comprising multiple bundles, rendering the simple diffusion tensor model unsuitable. Multi-compartment models deliver a convenient parameterization of the underlying complex fiber architecture, but pose challenges for fitting and model selection. Spherical deconvolution, in contrast, very economically produces a fiber orientation density function (fODF) without any explicit model assumptions. Since, however, the fODF is represented by spherical harmonics, a direct interpretation of the model parameters is impossible. Based on the fact that the fODF can often be interpreted as superposition of multiple peaks, each associated to one relatively coherent fiber population (bundle), we offer a solution that seeks to combine the advantages of both approaches: first the fiber configuration is modeled as fODF represented by spherical harmonics and then each of the peaks is parameterized separately in order to characterize the underlying bundle. In this work, the fODF peaks are approximated by Bingham distributions, capturing first and second-order statistics of the fiber orientations, from which we derive metrics for the parametric quantification of fiber bundles. We propose meaningful relationships between these measures and the underlying microstructural properties. We focus on metrics derived directly from properties of the Bingham distribution, such as peak length, peak direction, peak spread, integral over the peak, as well as a metric derived from the comparison of the largest peaks, which probes the complexity of the underlying microstructure. We compare these metrics to the conventionally used fractional anisotropy (FA) and show how they may help to increase the specificity of the characterization of microstructural properties. While metrics relying on the first moments of the Bingham distributions provide relatively robust results, second-order metrics representing the peak spread are only meaningful, if the SNR is very high and no fiber crossings are present in the voxel.

The Gini-coefficient: A new method to assess fetal brain development

Founded in 1963 The European Society of Paediatric Radiology 50th Annual Meeting and 36th Postgraduate Course of the European Society of Paediatric Radiology