Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

FIX/EHN update load_confound_strategy defaults #4225

Merged
merged 8 commits into from
Jan 19, 2024
+53 −6
Merged

FIX/EHN update load_confound_strategy defaults #4225

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
661c2f7
FIX/EHN update load_confound_strategy defaults
htwangtw Jan 16, 2024
b13c399
ADD Test for compcor global signal
htwangtw Jan 16, 2024
84dc583
update the example to reflect on fMRIPrep default parameters
htwangtw Jan 16, 2024
c6287b3
update release notes; no idea what's going on
htwangtw Jan 16, 2024
7c1f26f
Apply suggestions from code review
htwangtw Jan 17, 2024
b2da0f9
EHN stack level 2 makes more sense. only show warning when scrub is i…
htwangtw Jan 17, 2024
d41cccc
Update nilearn/interfaces/fmriprep/load_confounds_strategy.py
Remi-Gau Jan 18, 2024
2e0fdbd
Update doc/changes/latest.rst
Remi-Gau Jan 18, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
2 changes: 2 additions & 0 deletions doc/changes/latest.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,6 +40,7 @@ Enhancements
- :bdg-primary:`Doc` Specify full form of LSS in :ref:`sphx_glr_auto_examples_07_advanced_plot_beta_series.py` (:gh:`4141` by `Tamer Gezici`_).
- :bdg-primary:`Doc` Don't fetch tmaps in examples if tmaps aren't used in the example. (:gh:`4136` by `Christina Roßmanith`_).
- :bdg-primary:`Doc` Describe the return value in :func:`~nilearn.datasets.fetch_abide_pcp` documentation (:gh:`4159` by `Suramya Pokharel`_).
- :bdg-dark:`Code` Allow ``global_signal`` parameter in :func:`~interfaces.fmriprep.load_confounds_strategy` in ``denoise_strategy='compcor'`` (:gh:`4225` by `Hao-Ting Wang`_).

Changes
-------
Expand All @@ -55,3 +56,4 @@ Changes
- :bdg-dark:`Code` Remove leading underscore from non private functions to align with PEP8 (:gh:`4086` by `Rémi Gau`_).
- :bdg-dark:`Code` Make ``decoding/proximal_operator`` explicitly private to align with PEP8 (:gh:`4153` by `Rémi Gau`_).
- :bdg-dark:`Code` Make private functions public when used outside of their module ``nilearn.interface`` to align with PEP8 (:gh:`4168` by `Rémi Gau`_).
- :bdg-danger:`Deprecation` the default parameter of ``fd_threshold`` and ``std_dvars_threshold`` in :func:`~interfaces.fmriprep.load_confounds` with be updated to match ``fmriprep`` default in version 0.13 (:gh:`4225` by `Hao-Ting Wang`_).
4 changes: 2 additions & 2 deletions examples/03_connectivity/plot_signal_extraction.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -187,8 +187,8 @@
motion="basic",
wm_csf="basic",
scrub=5,
fd_threshold=0.2,
std_dvars_threshold=3,
fd_threshold=0.5,
std_dvars_threshold=1.5,
)

print(
Expand Down
34 changes: 33 additions & 1 deletion nilearn/interfaces/fmriprep/load_confounds.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -206,10 +206,19 @@ def load_confounds(
DVARS only.

fd_threshold : :obj:`float`, default=0.2

.. deprecated:: 0.11.0
The default value will be changed to 0.5 in 0.13.0

Framewise displacement threshold for scrub in mm.

std_dvars_threshold : :obj:`float`, default=3

.. deprecated:: 0.11.0
The default value will be changed to 1.5 in 0.13.0

Standardized DVARS threshold for scrub.
The default threshold matching :term:`fMRIPrep`.
DVARs is defined as root mean squared intensity difference of volume N
to volume N+1 :footcite:t:`Power2012`.
D referring to temporal derivative of timecourses,
Expand Down Expand Up @@ -295,7 +304,30 @@ def load_confounds(

"""
_check_strategy(strategy)

if "scrub" in strategy and fd_threshold == 0.2:
fd_threshold_default = (
"The default parameter for fd_threshold is currently 0.2 "
"which is inconsistent with the fMRIPrep default of 0.5. "
"In release 0.13.0, "
"the default strategy will be replaced by 0.5."
)
warnings.warn(
category=DeprecationWarning,
message=fd_threshold_default,
stacklevel=2,
)
if "scrub" in strategy and std_dvars_threshold == 3:
std_dvars_threshold_default = (
"The default parameter for std_dvars_threshold is currently 3 "
"which is inconsistent with the fMRIPrep default of 1.5. "
"In release 0.13.0, "
"the default strategy will be replaced by 1.5."
)
warnings.warn(
category=DeprecationWarning,
message=std_dvars_threshold_default,
stacklevel=2,
)
# load confounds per image provided
img_files, flag_single = sanitize_confounds(img_files)
confounds_out = []
Expand Down
11 changes: 8 additions & 3 deletions nilearn/interfaces/fmriprep/load_confounds_strategy.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,8 +27,8 @@
"motion": "full",
"wm_csf": "full",
"scrub": 5,
"fd_threshold": 0.2,
"std_dvars_threshold": 3,
"fd_threshold": 0.2, # updated here and doc to 0.5 in v0.13
"std_dvars_threshold": 3, # updated here and doc to 1.5 in v0.13
"global_signal": None,
"demean": True,
},
Expand All @@ -37,6 +37,7 @@
"motion": "full",
"n_compcor": "all",
"compcor": "anat_combined",
"global_signal": None,
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do we want to add this in an example?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

In fact we cannot create an example with the example dataset in the same tutorial because there's no json file for the confounds. We will have to update the example dataset to do so

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think this the 3rd instance where we cannot show things in example because we rely on too old datasets. Probably worth opening an issue to identify what BIDS dataset we have could potentially be updated.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sounds good to me. I did downsampled two fMRIPrep preprocessed dataset for my other projects as test data (ds000228, ds000030).

"demean": True,
},
"ica_aroma": {
Expand Down Expand Up @@ -101,6 +102,10 @@ def load_confounds_strategy(img_files, denoise_strategy="simple", **kwargs):
variance as the noise component estimation as the number of compcor
component can be really high. Please refer to :term:`fMRIPrep`
documentation for more details.

.. versionadded:: 0.11.0
`golobal_signal` is now a tunable parameter for compcor.

- 'ica_aroma': Load confounds for non-aggresive ICA-AROMA strategy
described in :footcite:t:`Pruim2015`. The strategy requires
:term:`fMRIPrep` outputs generated with `--use-aroma` suffixed with
Expand Down Expand Up @@ -152,7 +157,7 @@ def load_confounds_strategy(img_files, denoise_strategy="simple", **kwargs):
N/A N/A N/A N/A True*
scrubbing True full* full None* 5* 0.2* \
3* N/A N/A N/A True*
compcor True full* N/A N/A N/A N/A \
compcor True full* N/A None* N/A N/A \
N/A anat_combined* all* N/A True*
ica_aroma True N/A basic* None* N/A N/A \
N/A N/A N/A full True*
Expand Down
8 changes: 8 additions & 0 deletions nilearn/interfaces/fmriprep/tests/test_load_confounds_strategy.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -166,6 +166,14 @@ def test_strategy_compcor(tmp_path, fmriprep_version):
"a_comp_cor_57" if fmriprep_version == "1.4.x" else "w_comp_cor_00"
)
assert expected not in compcor_col_str_anat
assert "global_signal" not in compcor_col_str_anat

# test the global signal option
confounds, _ = load_confounds_strategy(
file_nii, denoise_strategy="compcor", global_signal="basic"
)
compcor_col_str_anat = "".join(confounds.columns)
assert "global_signal" in compcor_col_str_anat


@pytest.mark.parametrize("fmriprep_version", ["1.4.x", "21.x.x"])
Expand Down