[Neuroimaging] [DIPY] track simple (not DWI) vector field np.ndarray data

Francesco Giardini giardini at lens.unifi.it
Thu Jul 18 10:36:32 EDT 2019

I'm a complete novice at the field of Dipy so, first thank you very much to
all of you.
I would like to exploit Dipy tractography tools to draw my own data (not
DWI), but I have some problem about the fiber directions representation.

I work with 3D images of heart muscle tissue, and I developed a simple code
that extract the fibers orientation by means of a Structure Tensor

So, I have a 3d matrix 'R' (numpy.ndarray); every R(z, y, x) element
- three eigenvalues (3x1 matrx)
- three R3 eigenvectors (3x3 matrix)
- others informations (for example, the Fractional Anisotropy, useful (i
think) for the stop criterium)

In the attached, there is an example (only one XY plane displayed) of the
orientations maps, plotted with mayavi.

I'm trying to understand how to format my data to generate the streamlines.

I have reproduced some examples.
Here (
i find that csd_peaks contains the vectors informations (the source of the
tractography process?)

while here (
seems that tenfit contains the same vector field.

I'm reading also the documentation about dipy.tracking.local.LocalTracking
and dipy.tracking.local.DirectionGetter.

The question is: How I can create the streamlines from my data

I'am looking forward to your reply. Thank you very much.

Dr. Francesco Giardini (giardini at lens.unifi.it)

LENS - European Laboratory for Non-linear Spectroscopy
Via Nello Carrara 1, 50019 Sesto Fiorentino (FI), Italy
University of Florence - Biomedical Engineering

Tel. 055 457 2477 (off.)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20190718/647a38fd/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 3d_vectors_small.jpg
Type: image/jpeg
Size: 40250 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20190718/647a38fd/attachment.jpg>

More information about the Neuroimaging mailing list