ADNI T1 preprocessing: adni-to-bids followed by t1-linear, double bias field correction?

[Pauline Amrouche]

Hello,

We are preprocessing T1 ADNI images using Clinica.

It is mentionned in the adni-to-bids documentation that preprocessed N3(m) bias field corrected scans are selected when available for ADNI1,2,GO. However when using the t1-linear pipeline to preprocess the T1 images then N4ITK bias field correction is applied. Is there a risk of overcorrection in applying two successive bias field correction methods to the data? If yes, then would you recommend using non-N3(m) corrected scans for ADNI1,2,GO?

Thanks,

Pauline

[Alice Joubert]

Hello Pauline,

The N4ITK algorithm should correct the bias field only when needed. That means if the images were already corrected once, N4ITK could maybe apply a small correction but it should not lead to errors/overcorrection.

Best,

Alice

[Pauline Amrouche]

Hello Alice,

Thanks a lot for the answer.

Best,

Pauline