Getting Connectivity Matrix from Tractoflow

1

Hi there.

I am trying to estimate connectivity matrix from tractoflow output *.trk files.

I have following questions:

  1. What are the options that can used while running tractoflow to generate connectivity matrix in outputs
  2. If not generated in tractoflow outputs, whats the best way to generate it. I have used connectivitymatrix from dipy.tracking.utils. Any must todos while doing this? I get non-zero diagonal terms - why?
2

Hi @prabhjotkaur,

Tractoflow does not generate connectivity matrices.

There is no “best” way to generate it, and there are many options for atlases, streamline-to-region determination mechanisms, and edge weight normalization, each with their own biases. Without knowing more about your data and research questions, it would be difficult to prescribe an exact solution.

Some further reading:
https://www.sciencedirect.com/science/article/pii/S105381191930388X
https://onlinelibrary.wiley.com/doi/full/10.1002/jmri.27188
https://www.sciencedirect.com/science/article/pii/S1053811914003541
https://www.sciencedirect.com/science/article/pii/S1053811914008155
https://www.sciencedirect.com/science/article/pii/S1053811919310341

Streamlines that begin and end in the same region, or even just pass through the seed region while traveling, depending on how you constructed the matrix…

Best,
Steven

1 Like
3

Hi @prabhjotkaur ,

If you want to keep working in the tractoflow ecosystem, we developped connectoflow. Tomorrow, I’ll release the new container the corresponding container. You also can use combine_flows to organise your data accordingly. There is a PR specifically for connectoflow.

Let me know if it answers your questions.

Best,
Arnaud

1 Like
4

What are the input needs of connectflow?

Which atlas space does tractoflow and connectflow use? Does connectflow has option to provide aprac.aseg.nii.gz in other space?