ANTsR
ANTsR copied to clipboard
Subtraction methods fail in aslAveraging
Hi ANTsR team,
Describe the bug I'm trying the ASL processing tools available in ANTsR. Particularly to obtain pair-wise differences of a timeseries. I want to compare the different perfusion subtraction methods available. However, all four: "simpleSubtract","surroundSubtract","cubicSubtract" and ""sincSubtract" all crash with the following error message (example for one method):
ITK ExceptionObject caught! Error in aslAveraging(input_asl, mask = NULL, method = "surroundSubtract") : /usr/local/lib/R/site-library/ITKR/libs/include/ITK-5.1/itkImageSource.hxx:268: itk::ERROR: AverageOverDimensionImageFilter(0x563b5e72cf10): Subclass should override this method!!! If old behavior is desired invoke this->DynamicMultiThreadingOff(); before Update() is called. The best place is in class constructor.
To Reproduce Steps to reproduce the behavior: input_asl <- antsImageRead("test_asl_timeseries.nii.gz") perf_asl_surrsub <- aslAveraging(input_asl, mask = NULL, method = "surroundSubtract")
Expected behavior I expect the command to provide the temporal mean of the pair-wise difference of the ASL time-series. Isn't this right? Although, I'd prefer it give me the entire pair-wise difference time-series (before averaging) as well.
Desktop (please complete the following information):
- OS: Ubuntu
- Browser: Chrome
- Version: 18.04.4
Additional context Rstudio v 1.2.5001 and R version 3.6.3
Help is appreciated.
Thanks, Sri
@jeffduda any chance. you could take a look at this?
This appears to be an ANTs problem that is manifesting in ANTsR. It should just be a one line fix. I can fix it and push it if you want, or would you prefer a PR? Obviously, this fix won't filter up to ANTsR until the ANTs dependency is updated to a version with the fix. Not sure what the current policy is regarding how often the version of ANTs used by ANTsR is updated.
On Thu, Jul 16, 2020 at 7:50 AM stnava [email protected] wrote:
@jeffduda https://github.com/jeffduda any chance. you could take a look at this?
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659359993, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAD5NCFD6TUTQMW4RMPUQMTR33SPDANCNFSM4NMPEKJA .
No need for PR.
We are happy to do updates to tags as regularly as needed to fix bugs.
On Thu, Jul 16, 2020 at 5:47 PM Jeffrey Duda [email protected] wrote:
This appears to be an ANTs problem that is manifesting in ANTsR. It should just be a one line fix. I can fix it and push it if you want, or would you prefer a PR? Obviously, this fix won't filter up to ANTsR until the ANTs dependency is updated to a version with the fix. Not sure what the current policy is regarding how often the version of ANTs used by ANTsR is updated.
On Thu, Jul 16, 2020 at 7:50 AM stnava [email protected] wrote:
@jeffduda https://github.com/jeffduda any chance. you could take a look at this?
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659359993, or unsubscribe < https://github.com/notifications/unsubscribe-auth/AAD5NCFD6TUTQMW4RMPUQMTR33SPDANCNFSM4NMPEKJA
.
— You are receiving this because you commented.
Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659692254, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACPE7X5ZSUDFURN6WY3XU3R35YNPANCNFSM4NMPEKJA .
--
brian
Ok, I pushed the fix to ANTs and all looks good. I don't have push permission for ANTsRCore, but all you need to do is update the configure file with the new ANTs commit tag. I'll put up a PR with that tomorrow if you haven't had a chance to get to that. No changes to ANTsR itself, but it will need to be re-compiled. -jeff
On Thu, Jul 16, 2020 at 5:59 PM stnava [email protected] wrote:
No need for PR.
We are happy to do updates to tags as regularly as needed to fix bugs.
On Thu, Jul 16, 2020 at 5:47 PM Jeffrey Duda [email protected] wrote:
This appears to be an ANTs problem that is manifesting in ANTsR. It should just be a one line fix. I can fix it and push it if you want, or would you prefer a PR? Obviously, this fix won't filter up to ANTsR until the ANTs dependency is updated to a version with the fix. Not sure what the current policy is regarding how often the version of ANTs used by ANTsR is updated.
On Thu, Jul 16, 2020 at 7:50 AM stnava [email protected] wrote:
@jeffduda https://github.com/jeffduda any chance. you could take a look at this?
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659359993, or unsubscribe <
https://github.com/notifications/unsubscribe-auth/AAD5NCFD6TUTQMW4RMPUQMTR33SPDANCNFSM4NMPEKJA
.
— You are receiving this because you commented.
Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659692254, or unsubscribe < https://github.com/notifications/unsubscribe-auth/AACPE7X5ZSUDFURN6WY3XU3R35YNPANCNFSM4NMPEKJA
.
--
brian
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659697642, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAD5NCFCYRHKPEPKHBPZUETR35Z3NANCNFSM4NMPEKJA .
many thanks - yes a PR to antsrcore would be helpful.
brian
On Thu, Jul 16, 2020 at 7:35 PM Jeffrey Duda [email protected] wrote:
Ok, I pushed the fix to ANTs and all looks good. I don't have push permission for ANTsRCore, but all you need to do is update the configure file with the new ANTs commit tag. I'll put up a PR with that tomorrow if you haven't had a chance to get to that. No changes to ANTsR itself, but it will need to be re-compiled. -jeff
On Thu, Jul 16, 2020 at 5:59 PM stnava [email protected] wrote:
No need for PR.
We are happy to do updates to tags as regularly as needed to fix bugs.
On Thu, Jul 16, 2020 at 5:47 PM Jeffrey Duda [email protected] wrote:
This appears to be an ANTs problem that is manifesting in ANTsR. It should just be a one line fix. I can fix it and push it if you want, or would you prefer a PR? Obviously, this fix won't filter up to ANTsR until the ANTs dependency is updated to a version with the fix. Not sure what the current policy is regarding how often the version of ANTs used by ANTsR is updated.
On Thu, Jul 16, 2020 at 7:50 AM stnava [email protected] wrote:
@jeffduda https://github.com/jeffduda any chance. you could take a look at this?
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659359993, or unsubscribe <
https://github.com/notifications/unsubscribe-auth/AAD5NCFD6TUTQMW4RMPUQMTR33SPDANCNFSM4NMPEKJA
.
— You are receiving this because you commented.
Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659692254, or unsubscribe <
https://github.com/notifications/unsubscribe-auth/AACPE7X5ZSUDFURN6WY3XU3R35YNPANCNFSM4NMPEKJA
.
--
brian
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659697642, or unsubscribe < https://github.com/notifications/unsubscribe-auth/AAD5NCFCYRHKPEPKHBPZUETR35Z3NANCNFSM4NMPEKJA
.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659734816, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACPE7QUHPG4KPIT2ZDFXN3R36FEHANCNFSM4NMPEKJA .
PR submitted
On Thu, Jul 16, 2020 at 7:40 PM stnava [email protected] wrote:
many thanks - yes a PR to antsrcore would be helpful.
brian
On Thu, Jul 16, 2020 at 7:35 PM Jeffrey Duda [email protected] wrote:
Ok, I pushed the fix to ANTs and all looks good. I don't have push permission for ANTsRCore, but all you need to do is update the configure file with the new ANTs commit tag. I'll put up a PR with that tomorrow if you haven't had a chance to get to that. No changes to ANTsR itself, but it will need to be re-compiled. -jeff
On Thu, Jul 16, 2020 at 5:59 PM stnava [email protected] wrote:
No need for PR.
We are happy to do updates to tags as regularly as needed to fix bugs.
On Thu, Jul 16, 2020 at 5:47 PM Jeffrey Duda <[email protected]
wrote:
This appears to be an ANTs problem that is manifesting in ANTsR. It should just be a one line fix. I can fix it and push it if you want, or would you prefer a PR? Obviously, this fix won't filter up to ANTsR until the ANTs dependency is updated to a version with the fix. Not sure what the current policy is regarding how often the version of ANTs used by ANTsR is updated.
On Thu, Jul 16, 2020 at 7:50 AM stnava [email protected] wrote:
@jeffduda https://github.com/jeffduda any chance. you could take a look at this?
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659359993 , or unsubscribe <
https://github.com/notifications/unsubscribe-auth/AAD5NCFD6TUTQMW4RMPUQMTR33SPDANCNFSM4NMPEKJA
.
— You are receiving this because you commented.
Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659692254, or unsubscribe <
https://github.com/notifications/unsubscribe-auth/AACPE7X5ZSUDFURN6WY3XU3R35YNPANCNFSM4NMPEKJA
.
--
brian
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659697642, or unsubscribe <
https://github.com/notifications/unsubscribe-auth/AAD5NCFCYRHKPEPKHBPZUETR35Z3NANCNFSM4NMPEKJA
.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659734816, or unsubscribe < https://github.com/notifications/unsubscribe-auth/AACPE7QUHPG4KPIT2ZDFXN3R36FEHANCNFSM4NMPEKJA
.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659736379, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAD5NCGKYYJ5TOZESXAJELTR36FV3ANCNFSM4NMPEKJA .
it would be pretty easy to implement the request of returning difference pairs (for relevant methods). That would be an ANTsR update. Should I add that in (in a backwards compatible way) ? -jeff
On Sun, Jul 19, 2020 at 4:09 PM Jeffrey Duda [email protected] wrote:
PR submitted
On Thu, Jul 16, 2020 at 7:40 PM stnava [email protected] wrote:
many thanks - yes a PR to antsrcore would be helpful.
brian
On Thu, Jul 16, 2020 at 7:35 PM Jeffrey Duda [email protected] wrote:
Ok, I pushed the fix to ANTs and all looks good. I don't have push permission for ANTsRCore, but all you need to do is update the configure file with the new ANTs commit tag. I'll put up a PR with that tomorrow if you haven't had a chance to get to that. No changes to ANTsR itself, but it will need to be re-compiled. -jeff
On Thu, Jul 16, 2020 at 5:59 PM stnava [email protected] wrote:
No need for PR.
We are happy to do updates to tags as regularly as needed to fix bugs.
On Thu, Jul 16, 2020 at 5:47 PM Jeffrey Duda < [email protected]> wrote:
This appears to be an ANTs problem that is manifesting in ANTsR. It should just be a one line fix. I can fix it and push it if you want, or would you prefer a PR? Obviously, this fix won't filter up to ANTsR until the ANTs dependency is updated to a version with the fix. Not sure what the current policy is regarding how often the version of ANTs used by ANTsR is updated.
On Thu, Jul 16, 2020 at 7:50 AM stnava [email protected] wrote:
@jeffduda https://github.com/jeffduda any chance. you could take a look at this?
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659359993 , or unsubscribe <
https://github.com/notifications/unsubscribe-auth/AAD5NCFD6TUTQMW4RMPUQMTR33SPDANCNFSM4NMPEKJA
.
— You are receiving this because you commented.
Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659692254, or unsubscribe <
https://github.com/notifications/unsubscribe-auth/AACPE7X5ZSUDFURN6WY3XU3R35YNPANCNFSM4NMPEKJA
.
--
brian
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659697642, or unsubscribe <
https://github.com/notifications/unsubscribe-auth/AAD5NCFCYRHKPEPKHBPZUETR35Z3NANCNFSM4NMPEKJA
.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659734816, or unsubscribe < https://github.com/notifications/unsubscribe-auth/AACPE7QUHPG4KPIT2ZDFXN3R36FEHANCNFSM4NMPEKJA
.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsR/issues/319#issuecomment-659736379, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAD5NCGKYYJ5TOZESXAJELTR36FV3ANCNFSM4NMPEKJA .
that sounds good - I just tested the surround subtract and it worked in the sense that it did not crash. will try some time later this month on real data.