dipy logo

Site Navigation

NIPY Community

Previous topic

dipy.tracking.distances

Next topic

dipy.tracking.learning

dipy.tracking.eudx

class dipy.tracking.eudx.EuDX(a, ind, seeds, odf_vertices, a_low=0.0239, step_sz=0.5, ang_thr=60.0, length_thr=0.0, total_weight=0.5, max_points=1000, affine=None)

Euler Delta Crossings

Generates tracks with termination criteria defined by a delta function [R19] and it has similarities with FACT algorithm [R20] and Basser’s method but uses trilinear interpolation.

Can be used with any reconstruction method as DTI, DSI, QBI, GQI which can calculate an orientation distribution function and find the local peaks of that function. For example a single tensor model can give you only one peak a dual tensor model 2 peaks and quantitative anisotropy method as used in GQI can give you 3,4,5 or even more peaks.

The parameters of the delta function are checking thresholds for the direction propagation magnitude and the angle of propagation.

A specific number of seeds is defined randomly and then the tracks are generated for that seed if the delta function returns true.

Trilinear interpolation is being used for defining the weights of the propagation.

Notes

The coordinate system of the tractography is that of native space of image coordinates not native space world coordinates therefore voxel size is always considered as having size (1,1,1). Therefore, the origin is at the center of the center of the first voxel of the volume and all i,j,k coordinates start from the center of the voxel they represent.

References

[R19](1, 2) Garyfallidis, Towards an accurate brain tractography, PhD thesis, University of Cambridge, 2012.
[R20](1, 2) Mori et al. Three-dimensional tracking of axonal projections in the brain by magnetic resonance imaging. Ann. Neurol. 1999.