# Problems with fibers' directions

**URL:** <https://discourse.slicer.org/t/problems-with-fibers-directions/10759>\
**Category:** SlicerDMRI\
**Created:** [March 20, 2020, 9:27am UTC](https://discourse.slicer.org/t/problems-with-fibers-directions/10759 "2020-03-20T09:27:34Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![doc-xie](https://avatars.discourse-cdn.com/v4/letter/d/c77e96/32.png) [@doc-xie](https://discourse.slicer.org/u/doc-xie)\
**Post date:** [March 20, 2020, 9:27am UTC](https://discourse.slicer.org/t/problems-with-fibers-directions/10759/1 "2020-03-20T09:27:34Z")

</div>

Version: 4.11.0-2020-02-26  
System: Mac  
MR scanner: GE, 3.0T  
Problem:  
The components of the diffusion data can be checked successfully in Volume module and the FA map can be derived with Diffusion Tensor Estimation. The orientation of the map is marked correctly with anterior to posterior. But after running the whole brain fibers tractography with UKF method, the orientation changed into the directions (posterior to anterior) and contrary to the FA map. But if the data was loaded into Slicer using Dcm2niixGUI, the result was correct. What the reason about this and how to solve this problem?

 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/2/6/262e0437aabbdad0df2329fc86d854267043ceb7.jpeg)

 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/e/1/e19b4f58ae436f9548e17a53687d207cd329fa85.jpeg)  
 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/0/7/077c6003e10567ff731763eba376668b187bee0d.jpeg)  
the information after UKF tractography showed as below:  
UKF Tractography standard output:

Using the 1T simple model. Setting the default parameters accordingly:  
“\*”: set by user  
“-”: default setting

- stoppingFA: 0.06
- seedingThreshold: 0.1

- Qm: 0.005
- Ql: 300
- Rs: 0.01

- stepLength: 0.3
- recordLength: 0.9
- stoppingThreshold: 0.06
- seedsPerVoxel: 2  
Found 12 cores on your system.  
Running tractography with 12 thread(s).  
Number of non-zero gradients: 25  
Number of zero gradients: 1  
Permuting the axis order to: 0 1 2 3  
Resizing the data to: 25 256 256 39  
Computing the baseline image  
Dividing the signal by baseline image  
Data normalization finished!

Using 1-tensor simple model.  
Branching disabled

Using unconstrained filter  
nmse\_avg=0  
vtkDebugLeaks has found no leaks.

UKF Tractography completed without errors  
Very interesting, for another subject, after the diffusion data was loaded into Slicer with Dcm2niiGUI, the FA map looks normal but the fibers were in wrong direction again. If I handle the data with DTIPrep for motion and eddy current correction and load the data again, the fibers turned up down and only showed the superior part of the whole brain fibers.

 ![image](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/1/4/14091ffe885e3b495fb1bca2deae33911b5f30f5.jpeg) The information of UKF Tractography showed as below:  
UKF Tractography standard output:

Using the 1T simple model. Setting the default parameters accordingly:  
“\*”: set by user  
“-”: default setting

- stoppingFA: 0.15

- seedingThreshold: 0.18

- Qm: 0.005
- Ql: 300
- Rs: 0.01

- stepLength: 0.3
- recordLength: 0.9
- stoppingThreshold: 0.1
- seedsPerVoxel: 2  
Found 8 cores on your system.  
Running tractography with 8 thread(s).  
Number of non-zero gradients: 25  
Number of zero gradients: 1  
Permuting the axis order to: 0 1 2 3  
Resizing the data to: 25 256 256 40  
Computing the baseline image  
Dividing the signal by baseline image  
Data normalization finished!

Using 1-tensor simple model.  
Branching disabled

Using unconstrained filter  
nmse\_avg=0  
vtkDebugLeaks has found no leaks.

---

<div class="post-metadata">

**Author:** ![zhangfanmark](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/zhangfanmark/32/4451_2.png) [@zhangfanmark](https://discourse.slicer.org/u/zhangfanmark)\
**Post date:** [March 20, 2020, 1:22pm UTC](https://discourse.slicer.org/t/problems-with-fibers-directions/10759/2 "2020-03-20T13:22:55Z")

</div>

Hi @doc-xie

Given our conversation offline, I thought we have managed to run UKF in an appropriate way on these datasets. For the two screenshots you attached here, the tractography datasets were all visually wrong. This is related to the gradient table that is not in the format that SlicerDMRI wants.

Regards,  
Fan

---

<div class="post-metadata">

**Author:** ![doc-xie](https://avatars.discourse-cdn.com/v4/letter/d/c77e96/32.png) [@doc-xie](https://discourse.slicer.org/u/doc-xie)\
**Post date:** [March 25, 2020, 9:00am UTC](https://discourse.slicer.org/t/problems-with-fibers-directions/10759/3 "2020-03-25T09:00:48Z")

</div>

Maybe the reason about this problem is the version of Slicer. After I changed the software into 4.10.0, the direction of the fibers showed normally.  
Thank you very much.  
Xie