We suppose that the quantitative diffusion values FA, MD, AD & RD are calculated after the (automated) application of a smoothing filter. What kind of filter is used? Thank you in advance for your kind response.

Hello

There is a priori non smoothing filter applied before the computation of the MD, FA and other maps. The only smoothing filter that could exist is in the estimation in the tensors (the check boxes where you can choose to have smoothing when estimating tensors), and this one is a log-Euclidean anisotropic smoothing filter on tensors.

Could you tell me more on what makes you think there is a filter applied ?

Thanks

Thank you very much for your rapid reply!

The images visualised (FA, ADC, RD, AD) seem smoothed (see attachment), but it might be due to interpolation (acquired voxel size 1.4 x 1.4 x 3 mm3). Could you tell me what's the interpolation technique applied?
And just a last question: are the quantitative values calculated before or after interpolation?

Thank you very much! We really appreciate your help!

Hello again.

So from my understanding, in fact this is a visualization artefact. When we visualize any image in medinria, we use VTK and include a smoothing of the voxels at the time of visualization (no modification of the input image). This provides smoother contours generally but in some cases may seem weird especially if your image has large voxels.

First the answer: no we do not apply any smoothing in the processing of the data, meaning that if you export your FA generated in medinria, it will be fine. The easy solution to remove this visualization effect is just to click on the view of interest and type n .

Tell me if that helped or if I misunderstood your problem.

Great! Thank you very much for your adequate answers!

We appreciate your support!