help
help > RE: problem with functional outlier detection
Jul 8, 2020 12:07 PM | Alfonso Nieto-Castanon - Boston University
RE: problem with functional outlier detection
Dear Ilka,
The problem is likely a mis-interpretation of the format of the realignment parameter file. Currently CONN is able to interpret the following formats:
SPM rp_*.txt files: x y z (mm) pitch roll yaw (radians)
FSL *.par files: pitch roll yaw (radians) x y z (mm)
HCP *.deg.txt files: x y z (mm) pitch roll yaw (degrees)
Siemens *.siemens.txt files: y x -z (mm) roll pitch yaw (degrees)
Could you please check the documentation of the software you used for realignment to figure out the format of its output realignment parameter files? If it matches one of the above simply rename those realignment files to match the convention above so CONN is able to correctly interpret the format of the file. If it does not please let me know and I will add support for that format in the next release
Best
Alfonso
Originally posted by boehm:
The problem is likely a mis-interpretation of the format of the realignment parameter file. Currently CONN is able to interpret the following formats:
SPM rp_*.txt files: x y z (mm) pitch roll yaw (radians)
FSL *.par files: pitch roll yaw (radians) x y z (mm)
HCP *.deg.txt files: x y z (mm) pitch roll yaw (degrees)
Siemens *.siemens.txt files: y x -z (mm) roll pitch yaw (degrees)
Could you please check the documentation of the software you used for realignment to figure out the format of its output realignment parameter files? If it matches one of the above simply rename those realignment files to match the convention above so CONN is able to correctly interpret the format of the file. If it does not please let me know and I will add support for that format in the next release
Best
Alfonso
Originally posted by boehm:
Dear Conn-Team,
The resting state data I analysing using Conn consists of 2 sessions (recorded on different days). The Problem I encounter is that when I run the outlier detection within the preprocessing step in conn every volume is detected as an outlier in every participant. (of note my preprocessing is done outside of conn so I only do outliere detection in conn in order to be able to do scrubbing)
when ruminating about this Problem I came across some questions:
1. CONN attaches the two sessions in the art_screenshot (see attached). is that a problem (note the participant has left the scanner between the two session)
2. Is the outlier detection based on frame-wise displacement?
3. We are using the SpaceTimeRealign algorithm [Roche2011] (the reading of the realignment files seemed to have worked fine) .... but to make sure this is how a part of the
realignment parameter files look like
-0.0000000000 0.0000000000 0.0000000000 -0.0000000000 0.0000000000 -0.0000000000
0.0372224205 0.0133710840 -0.0001266183 0.0007184454 -0.0002899682 0.0000019261
-0.0339213207 -0.0016000893 -0.0161957437 -0.0011051932 0.0009239895 0.0000964644
-0.0220374585 -0.1113202351 -0.0511013495 -0.0049295098 0.0006219891 0.0002790727
Thanks for your help
Ilka
The resting state data I analysing using Conn consists of 2 sessions (recorded on different days). The Problem I encounter is that when I run the outlier detection within the preprocessing step in conn every volume is detected as an outlier in every participant. (of note my preprocessing is done outside of conn so I only do outliere detection in conn in order to be able to do scrubbing)
when ruminating about this Problem I came across some questions:
1. CONN attaches the two sessions in the art_screenshot (see attached). is that a problem (note the participant has left the scanner between the two session)
2. Is the outlier detection based on frame-wise displacement?
3. We are using the SpaceTimeRealign algorithm [Roche2011] (the reading of the realignment files seemed to have worked fine) .... but to make sure this is how a part of the
realignment parameter files look like
-0.0000000000 0.0000000000 0.0000000000 -0.0000000000 0.0000000000 -0.0000000000
0.0372224205 0.0133710840 -0.0001266183 0.0007184454 -0.0002899682 0.0000019261
-0.0339213207 -0.0016000893 -0.0161957437 -0.0011051932 0.0009239895 0.0000964644
-0.0220374585 -0.1113202351 -0.0511013495 -0.0049295098 0.0006219891 0.0002790727
Thanks for your help
Ilka
Threaded View
| Title | Author | Date |
|---|---|---|
| boehm | Jul 8, 2020 | |
| Alfonso Nieto-Castanon | Jul 8, 2020 | |
| boehm | Sep 1, 2020 | |
| boehm | Sep 2, 2020 | |