Co registration issues with derivates provided

Co registration issues with derivates provided  

  By: eurunuela on July 15, 2024, 9:44 p.m.

Hello,

I don't know if you are aware of this but it seems like some of the preprocessed CTP data provided may have not been properly preprocessed. I have seen slices where it is very clear that the co-registration step may have gone wrong. See the example below:

 Last edited by: eurunuela on July 16, 2024, 3:18 p.m., edited 1 time in total.

Re: Skull stripping registration issues with derivates provided  

  By: ezequieldlrosa on July 15, 2024, 10:28 p.m.

Hi Eneko,

We are aware of this, and it is (somehow) expected.

It is not a co-registration issue, but the 'defacing' result aimed to de-identify patients (a step requested by the centers providing the data). Face bones, muscles and skin were the target. For NCCT and CTA, since the face region is mostly available, the removed (defaced) areas are more symmetrical; however, the defaced regions might be worse for CTP, since it has limited head coverage (not always including the face).

Just to let you know- face areas were removed while ensuring that the brain is preserved.

Hope this helps. Ezequiel

Re: Co registration issues with derivates provided  

  By: eurunuela on July 16, 2024, 3:18 p.m.

Thank you Ezequiel!