intensity histogram of b0 image after preprocessing

Submitted by lizhaoqing on

Dear Tortoise team,

I have some questions about the changes of histogram of b0 image after preprocessing including motion & eddy distortion correction, EPI correction (using DRBUDDI), gibbs ringing correction, denoising.

In the picture, it is shown that the histogram of b0 image of raw data (3800_b0); of b0 image generated after motion & eddy distortion correction, gibbs ringing correction, denoising (3800_proc_b0), and of b0 image generated after EPI distortion correction (3800_proc_DRBUDDI_proc_b0). It seems that the histogram has changes after preprocessing (using DIFFPREP and DRBUDDI). Is there general explanation for these changes? 


Looking forward to your reply. Thanks very much.


Submitted by irfanogl on Mon, 11/30/2020 - 15:41



The histogram is expected to change because the signal is manipulated in many ways.

Denoising, gibbs ringing, eddy currents distortion correction (due to signal modulation with Jacobians),  EPI distortion correction (again Jacobians), + INTERPOLATION,  ALL change the signal.

So again this is expected and not a big concern.