help
help > RE: Regression and bandpass filtering in fMRI
Jun 27, 2020 02:06 PM | Alfonso Nieto-Castanon - Boston University
RE: Regression and bandpass filtering in fMRI
H Karel,
Yes, that is perfectly valid, you are simply performing a frequency-specific analysis and evaluating connectivity at the desired frequency band (of course, sensitivity might be low due to the reduced degrees of freedom of your first-level measures so you will need a considerable number of subjects to get a clear picture)
Best
Alfonso
Originally posted by Karel Mauricio López Vilaret:
Yes, that is perfectly valid, you are simply performing a frequency-specific analysis and evaluating connectivity at the desired frequency band (of course, sensitivity might be low due to the reduced degrees of freedom of your first-level measures so you will need a considerable number of subjects to get a clear picture)
Best
Alfonso
Originally posted by Karel Mauricio López Vilaret:
Hi Alfonso,
Thank you so much for your reply.
That make sense. In fact, after doing RegBP I did not have any problem. But, considering that bandpassing at these very low frequency ranges ("0.002 - 0.01" and "0.012 - 0.028") removes the majority of the independent components of the signal, will the resulting output has any statistical validity?
Thanks in advance.
Best regards,
Karel
Thank you so much for your reply.
That make sense. In fact, after doing RegBP I did not have any problem. But, considering that bandpassing at these very low frequency ranges ("0.002 - 0.01" and "0.012 - 0.028") removes the majority of the independent components of the signal, will the resulting output has any statistical validity?
Thanks in advance.
Best regards,
Karel
Threaded View
| Title | Author | Date |
|---|---|---|
| Karel Mauricio López Vilaret | Jun 16, 2020 | |
| Alfonso Nieto-Castanon | Jun 18, 2020 | |
| Karel Mauricio López Vilaret | Jun 24, 2020 | |
| Alfonso Nieto-Castanon | Jun 27, 2020 | |