Partial volume effect and MAP estimation

SCT
1

I use MAP method for metrics’ extraction, so for each subject I have an average and a standard deviation for each metric in each ROI. My question is about how to use the std associated with each subject: how can I account for it if I wanted to compute also the general std of averages across subjects?
Please find attached an example file for all my subjects and FA metric in wm ROI. dki_FA_wm.csv (2.1 KB)
Hope to be enough clear

thank you in advance,

Rosella Trò

2

Hi @Rosella_Tro

Thank you for reaching out.

I use MAP method for metrics’ extraction, so for each subject I have an average and a standard deviation for each metric in each ROI.

To be more precise, if you use the “MAP” method (-method map), you would obtain the MAP, not the average. The average would be obtained with wa or with bin.

However, in the CSV file you sent, it shows “WA()” (not “MAP()”), suggesting that you did not use -method map. To avoid confusion in the future, it would be better if you could join the script or exact syntax you ran.

My question is about how to use the std associated with each subject: how can I account for it if I wanted to compute also the general std of averages across subjects?

It depends what you would like to report. E.g.,

  • if you would like to report the average of the STD (ie intra-patient variability), then you can average the STD across subjects.
  • If you would like to report the inter-subject STD, then you can compute the STD, across subjects, of the intra-subject average.

Best,
Julien

3

Hi,
sorry I finally chose to use weighted average method instead of MAP. However, the question also stands for wa method and I think your answer gave me the solution.

1 Like