Extracting signals from a brain parcellation

Here we show how to extract signals from a brain parcellation and compute a correlation matrix.

We also show the importance of defining good confounds signals: the first correlation matrix is computed after regressing out simple confounds signals: movement regressors, white matter and CSF signals, … The second one is without any confounds: all regions are connected to each other. Finally we demonstrated the functionality of nilearn.interfaces.fmriprep.load_confounds to flexibly select confound variables from fMRIPrep outputs while following some implementation guideline of fMRIPrep confounds documentation https://fmriprep.org/en/stable/outputs.html#confounds.

One reference that discusses the importance of confounds is Varoquaux and Craddock[1].

This is just a code example, see the corresponding section in the documentation for more.

Note

This example needs SciPy >= 1.0.0 for the reordering of the matrix.

Note

If you are using Nilearn with a version older than 0.9.0, then you should either upgrade your version or import maskers from the input_data module instead of the maskers module.

That is, you should manually replace in the following example all occurrences of:

from nilearn.maskers import NiftiMasker

with:

from nilearn.input_data import NiftiMasker

Retrieve the atlas and the data

from nilearn import datasets

dataset = datasets.fetch_atlas_harvard_oxford("cort-maxprob-thr25-2mm")
atlas_filename = dataset.maps
labels = dataset.labels

print(f"Atlas ROIs are located in nifti image (4D) at: {atlas_filename}")

# One subject of brain development fMRI data
data = datasets.fetch_development_fmri(n_subjects=1, reduce_confounds=True)
fmri_filenames = data.func[0]
reduced_confounds = data.confounds[0]  # This is a preselected set of confounds

Atlas ROIs are located in nifti image (4D) at:
<class 'nibabel.nifti1.Nifti1Image'>
data shape (91, 109, 91)
affine:
[[   2.    0.    0.  -90.]
 [   0.    2.    0. -126.]
 [   0.    0.    2.  -72.]
 [   0.    0.    0.    1.]]
metadata:
<class 'nibabel.nifti1.Nifti1Header'> object, endian='<'
sizeof_hdr      : 348
data_type       : b''
db_name         : b''
extents         : 0
session_error   : 0
regular         : b''
dim_info        : 0
dim             : [  3  91 109  91   1   1   1   1]
intent_p1       : 0.0
intent_p2       : 0.0
intent_p3       : 0.0
intent_code     : none
datatype        : uint8
bitpix          : 8
slice_start     : 0
pixdim          : [1. 2. 2. 2. 1. 1. 1. 1.]
vox_offset      : 0.0
scl_slope       : nan
scl_inter       : nan
slice_end       : 0
slice_code      : unknown
xyzt_units      : 0
cal_max         : 0.0
cal_min         : 0.0
slice_duration  : 0.0
toffset         : 0.0
glmax           : 0
glmin           : 0
descrip         : b''
aux_file        : b''
qform_code      : unknown
sform_code      : aligned
quatern_b       : 0.0
quatern_c       : 0.0
quatern_d       : 0.0
qoffset_x       : -90.0
qoffset_y       : -126.0
qoffset_z       : -72.0
srow_x          : [  2.   0.   0. -90.]
srow_y          : [   0.    2.    0. -126.]
srow_z          : [  0.   0.   2. -72.]
intent_name     : b''
magic           : b'n+1'

Extract signals on a parcellation defined by labels

Using the NiftiLabelsMasker

from nilearn.maskers import NiftiLabelsMasker

masker = NiftiLabelsMasker(
    labels_img=atlas_filename,
    standardize="zscore_sample",
    standardize_confounds="zscore_sample",
    memory="nilearn_cache",
    verbose=5,
)

# Here we go from nifti files to the signal time series in a numpy
# array. Note how we give confounds to be regressed out during signal
# extraction
time_series = masker.fit_transform(fmri_filenames, confounds=reduced_confounds)

[NiftiLabelsMasker.wrapped] loading data from Nifti1Image(
shape=(91, 109, 91),
affine=array([[   2.,    0.,    0.,  -90.],
       [   0.,    2.,    0., -126.],
       [   0.,    0.,    2.,  -72.],
       [   0.,    0.,    0.,    1.]])
)
Resampling labels
________________________________________________________________________________
[Memory] Calling nilearn.maskers.base_masker._filter_and_extract...
_filter_and_extract('/home/runner/work/nilearn/nilearn/nilearn_data/development_fmri/development_fmri/sub-pixar123_task-pixar_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz',
<nilearn.maskers.nifti_labels_masker._ExtractionFunctor object at 0x7f8529f91f70>,
{ 'background_label': 0,
  'clean_kwargs': {},
  'detrend': False,
  'dtype': None,
  'high_pass': None,
  'high_variance_confounds': False,
  'keep_masked_labels': True,
  'labels': None,
  'labels_img': <nibabel.nifti1.Nifti1Image object at 0x7f851abac5c0>,
  'low_pass': None,
  'mask_img': None,
  'reports': True,
  'smoothing_fwhm': None,
  'standardize': 'zscore_sample',
  'standardize_confounds': 'zscore_sample',
  'strategy': 'mean',
  't_r': None,
  'target_affine': None,
  'target_shape': None}, confounds=[ '/home/runner/work/nilearn/nilearn/nilearn_data/development_fmri/development_fmri/sub-pixar123_task-pixar_desc-reducedConfounds_regressors.tsv'], sample_mask=None, dtype=None, memory=Memory(location=nilearn_cache/joblib), memory_level=1, verbose=5)
[NiftiLabelsMasker.transform_single_imgs] Loading data from /home/runner/work/nilearn/nilearn/nilearn_data/development_fmri/development_fmri/sub-pixar123_task-pixar_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz
[NiftiLabelsMasker.transform_single_imgs] Extracting region signals
[NiftiLabelsMasker.transform_single_imgs] Cleaning extracted signals
_______________________________________________filter_and_extract - 0.9s, 0.0min

Compute and display a correlation matrix

from nilearn.connectome import ConnectivityMeasure

correlation_measure = ConnectivityMeasure(
    kind="correlation",
    standardize="zscore_sample",
)
correlation_matrix = correlation_measure.fit_transform([time_series])[0]

# Plot the correlation matrix
import numpy as np

from nilearn import plotting

# Make a large figure
# Mask the main diagonal for visualization:
np.fill_diagonal(correlation_matrix, 0)
# The labels we have start with the background (0), hence we skip the
# first label
# matrices are ordered for block-like representation
plotting.plot_matrix(
    correlation_matrix,
    figure=(10, 8),
    labels=labels[1:],
    vmax=0.8,
    vmin=-0.8,
    title="Confounds",
    reorder=True,
)

<matplotlib.image.AxesImage object at 0x7f85200e8110>

Extract signals and compute a connectivity matrix without confounds removal

After covering the basic of signal extraction and functional connectivity matrix presentation, let’s look into the impact of confounds to fMRI signal and functional connectivity. Firstly let’s find out what a functional connectivity matrix looks like without confound removal.

time_series = masker.fit_transform(fmri_filenames)
# Note how we did not specify confounds above. This is bad!

correlation_matrix = correlation_measure.fit_transform([time_series])[0]

np.fill_diagonal(correlation_matrix, 0)

plotting.plot_matrix(
    correlation_matrix,
    figure=(10, 8),
    labels=labels[1:],
    vmax=0.8,
    vmin=-0.8,
    title="No confounds",
    reorder=True,
)

[NiftiLabelsMasker.wrapped] loading data from Nifti1Image(
shape=(91, 109, 91),
affine=array([[   2.,    0.,    0.,  -90.],
       [   0.,    2.,    0., -126.],
       [   0.,    0.,    2.,  -72.],
       [   0.,    0.,    0.,    1.]])
)
________________________________________________________________________________
[Memory] Calling nilearn.maskers.base_masker._filter_and_extract...
_filter_and_extract('/home/runner/work/nilearn/nilearn/nilearn_data/development_fmri/development_fmri/sub-pixar123_task-pixar_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz',
<nilearn.maskers.nifti_labels_masker._ExtractionFunctor object at 0x7f85200e8920>,
{ 'background_label': 0,
  'clean_kwargs': {},
  'detrend': False,
  'dtype': None,
  'high_pass': None,
  'high_variance_confounds': False,
  'keep_masked_labels': True,
  'labels': None,
  'labels_img': <nibabel.nifti1.Nifti1Image object at 0x7f851abac5c0>,
  'low_pass': None,
  'mask_img': None,
  'reports': True,
  'smoothing_fwhm': None,
  'standardize': 'zscore_sample',
  'standardize_confounds': 'zscore_sample',
  'strategy': 'mean',
  't_r': None,
  'target_affine': None,
  'target_shape': None}, confounds=None, sample_mask=None, dtype=None, memory=Memory(location=nilearn_cache/joblib), memory_level=1, verbose=5)
[NiftiLabelsMasker.transform_single_imgs] Loading data from /home/runner/work/nilearn/nilearn/nilearn_data/development_fmri/development_fmri/sub-pixar123_task-pixar_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz
[NiftiLabelsMasker.transform_single_imgs] Extracting region signals
[NiftiLabelsMasker.transform_single_imgs] Cleaning extracted signals
_______________________________________________filter_and_extract - 0.9s, 0.0min

<matplotlib.image.AxesImage object at 0x7f8529f05730>

Load confounds from file using a flexible strategy with fmriprep interface

The nilearn.interfaces.fmriprep.load_confounds function provides flexible parameters to retrieve the relevant columns from the TSV file generated by fMRIPrep. nilearn.interfaces.fmriprep.load_confounds ensures two things:

1. The correct regressors are selected with provided strategy, and

2. Volumes such as non-steady-state and/or high motion volumes are masked out correctly.

Let’s try a simple strategy removing motion, white matter signal, cerebrospinal fluid signal with high-pass filtering.

from nilearn.interfaces.fmriprep import load_confounds

confounds_simple, sample_mask = load_confounds(
    fmri_filenames,
    strategy=["high_pass", "motion", "wm_csf"],
    motion="basic",
    wm_csf="basic",
)

print("The shape of the confounds matrix is:", confounds_simple.shape)
print(confounds_simple.columns)

time_series = masker.fit_transform(
    fmri_filenames, confounds=confounds_simple, sample_mask=sample_mask
)

correlation_matrix = correlation_measure.fit_transform([time_series])[0]

np.fill_diagonal(correlation_matrix, 0)

plotting.plot_matrix(
    correlation_matrix,
    figure=(10, 8),
    labels=labels[1:],
    vmax=0.8,
    vmin=-0.8,
    title="Motion, WM, CSF",
    reorder=True,
)

The shape of the confounds matrix is: (168, 12)
Index(['cosine00', 'cosine01', 'cosine02', 'cosine03', 'csf', 'rot_x', 'rot_y',
       'rot_z', 'trans_x', 'trans_y', 'trans_z', 'white_matter'],
      dtype='object')
[NiftiLabelsMasker.wrapped] loading data from Nifti1Image(
shape=(91, 109, 91),
affine=array([[   2.,    0.,    0.,  -90.],
       [   0.,    2.,    0., -126.],
       [   0.,    0.,    2.,  -72.],
       [   0.,    0.,    0.,    1.]])
)
________________________________________________________________________________
[Memory] Calling nilearn.maskers.base_masker._filter_and_extract...
_filter_and_extract('/home/runner/work/nilearn/nilearn/nilearn_data/development_fmri/development_fmri/sub-pixar123_task-pixar_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz',
<nilearn.maskers.nifti_labels_masker._ExtractionFunctor object at 0x7f851bbc8530>,
{ 'background_label': 0,
  'clean_kwargs': {},
  'detrend': False,
  'dtype': None,
  'high_pass': None,
  'high_variance_confounds': False,
  'keep_masked_labels': True,
  'labels': None,
  'labels_img': <nibabel.nifti1.Nifti1Image object at 0x7f851abac5c0>,
  'low_pass': None,
  'mask_img': None,
  'reports': True,
  'smoothing_fwhm': None,
  'standardize': 'zscore_sample',
  'standardize_confounds': 'zscore_sample',
  'strategy': 'mean',
  't_r': None,
  'target_affine': None,
  'target_shape': None}, confounds=[      cosine00  cosine01  cosine02  ...   trans_y   trans_z  white_matter
0    0.109104  0.109090  0.109066  ... -0.026078  0.055006     -0.876886
1    0.109066  0.108937  0.108723  ... -0.027587  0.049458     -1.418909
2    0.108990  0.108632  0.108038  ... -0.019085  0.075787     -1.540842
3    0.108875  0.108176  0.107012  ... -0.023900  0.053022     -1.922085
4    0.108723  0.107567  0.105651  ... -0.033396  0.077764     -1.843388
..        ...       ...       ...  ...       ...       ...           ...
163 -0.108723  0.107567 -0.105651  ...  0.064873 -0.022169      1.221217
164 -0.108875  0.108176 -0.107012  ... -0.064266 -0.042248      1.237783
165 -0.108990  0.108632 -0.108038  ...  0.053241 -0.029500      1.590905
166 -0.109066  0.108937 -0.108723  ... -0.081559 -0.034236      1.122246
167 -0.109104  0.109090 -0.109066  ...  0.069287 -0.007294      1.000828

[168 rows x 12 columns]], sample_mask=None, dtype=None, memory=Memory(location=nilearn_cache/joblib), memory_level=1, verbose=5)
[NiftiLabelsMasker.transform_single_imgs] Loading data from /home/runner/work/nilearn/nilearn/nilearn_data/development_fmri/development_fmri/sub-pixar123_task-pixar_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz
[NiftiLabelsMasker.transform_single_imgs] Extracting region signals
[NiftiLabelsMasker.transform_single_imgs] Cleaning extracted signals
_______________________________________________filter_and_extract - 0.9s, 0.0min

<matplotlib.image.AxesImage object at 0x7f8513d6dbb0>

Motion-based scrubbing

With a scrubbing-based strategy, load_confounds returns a sample_mask that removes the index of volumes exceeding the framewise displacement and standardised DVARS threshold, and all the continuous segment with less than five volumes. Before applying scrubbing, it’s important to access the percentage of volumns scrubbed. Scrubbing is not a suitable strategy for datasets with too many high motion subjects. On top of the simple strategy above, let’s add scrubbing to our strategy.

confounds_scrub, sample_mask = load_confounds(
    fmri_filenames,
    strategy=["high_pass", "motion", "wm_csf", "scrub"],
    motion="basic",
    wm_csf="basic",
    scrub=5,
    fd_threshold=0.5,
    std_dvars_threshold=1.5,
)

print(
    f"After scrubbing, {sample_mask.shape[0]} "
    f"out of {confounds_scrub.shape[0]} volumes remains"
)
print("The shape of the confounds matrix is:", confounds_simple.shape)
print(confounds_scrub.columns)

time_series = masker.fit_transform(
    fmri_filenames, confounds=confounds_scrub, sample_mask=sample_mask
)

correlation_matrix = correlation_measure.fit_transform([time_series])[0]

np.fill_diagonal(correlation_matrix, 0)

plotting.plot_matrix(
    correlation_matrix,
    figure=(10, 8),
    labels=labels[1:],
    vmax=0.8,
    vmin=-0.8,
    title="Motion, WM, CSF, Scrubbing",
    reorder=True,
)

After scrubbing, 164 out of 168 volumes remains
The shape of the confounds matrix is: (168, 12)
Index(['cosine00', 'cosine01', 'cosine02', 'cosine03', 'csf', 'rot_x', 'rot_y',
       'rot_z', 'trans_x', 'trans_y', 'trans_z', 'white_matter'],
      dtype='object')
[NiftiLabelsMasker.wrapped] loading data from Nifti1Image(
shape=(91, 109, 91),
affine=array([[   2.,    0.,    0.,  -90.],
       [   0.,    2.,    0., -126.],
       [   0.,    0.,    2.,  -72.],
       [   0.,    0.,    0.,    1.]])
)
________________________________________________________________________________
[Memory] Calling nilearn.maskers.base_masker._filter_and_extract...
_filter_and_extract('/home/runner/work/nilearn/nilearn/nilearn_data/development_fmri/development_fmri/sub-pixar123_task-pixar_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz',
<nilearn.maskers.nifti_labels_masker._ExtractionFunctor object at 0x7f8513cce930>,
{ 'background_label': 0,
  'clean_kwargs': {},
  'detrend': False,
  'dtype': None,
  'high_pass': None,
  'high_variance_confounds': False,
  'keep_masked_labels': True,
  'labels': None,
  'labels_img': <nibabel.nifti1.Nifti1Image object at 0x7f851abac5c0>,
  'low_pass': None,
  'mask_img': None,
  'reports': True,
  'smoothing_fwhm': None,
  'standardize': 'zscore_sample',
  'standardize_confounds': 'zscore_sample',
  'strategy': 'mean',
  't_r': None,
  'target_affine': None,
  'target_shape': None}, confounds=[      cosine00  cosine01  cosine02  ...   trans_y   trans_z  white_matter
0    0.108440  0.106895  0.110644  ... -0.026366  0.054759     -0.878390
1    0.108401  0.106742  0.110301  ... -0.027876  0.049211     -1.420413
2    0.108325  0.106438  0.109616  ... -0.019374  0.075540     -1.542346
3    0.108211  0.105981  0.108591  ... -0.024188  0.052775     -1.923589
4    0.108058  0.105373  0.107229  ... -0.033685  0.077517     -1.844892
..        ...       ...       ...  ...       ...       ...           ...
163 -0.109387  0.105373 -0.104073  ...  0.064585 -0.022416      1.219713
164 -0.109540  0.105981 -0.105434  ... -0.064554 -0.042495      1.236279
165 -0.109654  0.106438 -0.106460  ...  0.052952 -0.029747      1.589401
166 -0.109731  0.106742 -0.107145  ... -0.081847 -0.034483      1.120742
167 -0.109769  0.106895 -0.107488  ...  0.068998 -0.007541      0.999324

[168 rows x 12 columns]], sample_mask=array([  0, ..., 167]), dtype=None, memory=Memory(location=nilearn_cache/joblib), memory_level=1, verbose=5)
[NiftiLabelsMasker.transform_single_imgs] Loading data from /home/runner/work/nilearn/nilearn/nilearn_data/development_fmri/development_fmri/sub-pixar123_task-pixar_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz
[NiftiLabelsMasker.transform_single_imgs] Extracting region signals
[NiftiLabelsMasker.transform_single_imgs] Cleaning extracted signals
_______________________________________________filter_and_extract - 0.9s, 0.0min

<matplotlib.image.AxesImage object at 0x7f851d7026f0>

The impact of global signal removal

Global signal removes the grand mean from your signal. The benefit is that it can remove impacts of physiological artifacts with minimal impact on the degrees of freedom. The downside is that one cannot get insight into variance explained by certain sources of noise. Now let’s add global signal to the simple strategy and see its impact.

confounds_minimal_no_gsr, sample_mask = load_confounds(
    fmri_filenames,
    strategy=["high_pass", "motion", "wm_csf", "global_signal"],
    motion="basic",
    wm_csf="basic",
    global_signal="basic",
)
print("The shape of the confounds matrix is:", confounds_minimal_no_gsr.shape)
print(confounds_minimal_no_gsr.columns)

time_series = masker.fit_transform(
    fmri_filenames, confounds=confounds_minimal_no_gsr, sample_mask=sample_mask
)

correlation_matrix = correlation_measure.fit_transform([time_series])[0]

np.fill_diagonal(correlation_matrix, 0)

plotting.plot_matrix(
    correlation_matrix,
    figure=(10, 8),
    labels=labels[1:],
    vmax=0.8,
    vmin=-0.8,
    title="Motion, WM, CSF, GSR",
    reorder=True,
)

The shape of the confounds matrix is: (168, 13)
Index(['cosine00', 'cosine01', 'cosine02', 'cosine03', 'csf', 'global_signal',
       'rot_x', 'rot_y', 'rot_z', 'trans_x', 'trans_y', 'trans_z',
       'white_matter'],
      dtype='object')
[NiftiLabelsMasker.wrapped] loading data from Nifti1Image(
shape=(91, 109, 91),
affine=array([[   2.,    0.,    0.,  -90.],
       [   0.,    2.,    0., -126.],
       [   0.,    0.,    2.,  -72.],
       [   0.,    0.,    0.,    1.]])
)
________________________________________________________________________________
[Memory] Calling nilearn.maskers.base_masker._filter_and_extract...
_filter_and_extract('/home/runner/work/nilearn/nilearn/nilearn_data/development_fmri/development_fmri/sub-pixar123_task-pixar_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz',
<nilearn.maskers.nifti_labels_masker._ExtractionFunctor object at 0x7f8529f3b020>,
{ 'background_label': 0,
  'clean_kwargs': {},
  'detrend': False,
  'dtype': None,
  'high_pass': None,
  'high_variance_confounds': False,
  'keep_masked_labels': True,
  'labels': None,
  'labels_img': <nibabel.nifti1.Nifti1Image object at 0x7f851abac5c0>,
  'low_pass': None,
  'mask_img': None,
  'reports': True,
  'smoothing_fwhm': None,
  'standardize': 'zscore_sample',
  'standardize_confounds': 'zscore_sample',
  'strategy': 'mean',
  't_r': None,
  'target_affine': None,
  'target_shape': None}, confounds=[      cosine00  cosine01  cosine02  ...   trans_y   trans_z  white_matter
0    0.109104  0.109090  0.109066  ... -0.026078  0.055006     -0.876886
1    0.109066  0.108937  0.108723  ... -0.027587  0.049458     -1.418909
2    0.108990  0.108632  0.108038  ... -0.019085  0.075787     -1.540842
3    0.108875  0.108176  0.107012  ... -0.023900  0.053022     -1.922085
4    0.108723  0.107567  0.105651  ... -0.033396  0.077764     -1.843388
..        ...       ...       ...  ...       ...       ...           ...
163 -0.108723  0.107567 -0.105651  ...  0.064873 -0.022169      1.221217
164 -0.108875  0.108176 -0.107012  ... -0.064266 -0.042248      1.237783
165 -0.108990  0.108632 -0.108038  ...  0.053241 -0.029500      1.590905
166 -0.109066  0.108937 -0.108723  ... -0.081559 -0.034236      1.122246
167 -0.109104  0.109090 -0.109066  ...  0.069287 -0.007294      1.000828

[168 rows x 13 columns]], sample_mask=None, dtype=None, memory=Memory(location=nilearn_cache/joblib), memory_level=1, verbose=5)
[NiftiLabelsMasker.transform_single_imgs] Loading data from /home/runner/work/nilearn/nilearn/nilearn_data/development_fmri/development_fmri/sub-pixar123_task-pixar_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz
[NiftiLabelsMasker.transform_single_imgs] Extracting region signals
[NiftiLabelsMasker.transform_single_imgs] Cleaning extracted signals
_______________________________________________filter_and_extract - 0.9s, 0.0min

<matplotlib.image.AxesImage object at 0x7f8529e6f950>

Using predefined strategies

Instead of customising the strategy through nilearn.interfaces.fmriprep.load_confounds, one can use a predefined strategy with nilearn.interfaces.fmriprep.load_confounds_strategy. Based on the confound variables generated through fMRIPrep, and past benchmarks studies (Ciric et al.[2], Parkes et al.[3]): simple, scrubbing, compcor, ica_aroma. The following examples shows how to use the simple strategy and overwrite the motion default to basic.

from nilearn.interfaces.fmriprep import load_confounds_strategy

# use default parameters
confounds, sample_mask = load_confounds_strategy(
    fmri_filenames, denoise_strategy="simple", motion="basic"
)
time_series = masker.fit_transform(
    fmri_filenames, confounds=confounds, sample_mask=sample_mask
)

correlation_matrix = correlation_measure.fit_transform([time_series])[0]

np.fill_diagonal(correlation_matrix, 0)

plotting.plot_matrix(
    correlation_matrix,
    figure=(10, 8),
    labels=labels[1:],
    vmax=0.8,
    vmin=-0.8,
    title="simple",
    reorder=True,
)

# add optional parameter global signal
confounds, sample_mask = load_confounds_strategy(
    fmri_filenames,
    denoise_strategy="simple",
    motion="basic",
    global_signal="basic",
)
time_series = masker.fit_transform(
    fmri_filenames, confounds=confounds, sample_mask=sample_mask
)

correlation_matrix = correlation_measure.fit_transform([time_series])[0]

np.fill_diagonal(correlation_matrix, 0)

plotting.plot_matrix(
    correlation_matrix,
    figure=(10, 8),
    labels=labels[1:],
    vmax=0.8,
    vmin=-0.8,
    title="simple with global signal",
    reorder=True,
)

plotting.show()

• 
• 

[NiftiLabelsMasker.wrapped] loading data from Nifti1Image(
shape=(91, 109, 91),
affine=array([[   2.,    0.,    0.,  -90.],
       [   0.,    2.,    0., -126.],
       [   0.,    0.,    2.,  -72.],
       [   0.,    0.,    0.,    1.]])
)
[Memory]12.4s, 0.2min   : Loading _filter_and_extract...
__________________________________filter_and_extract cache loaded - 0.0s, 0.0min
[NiftiLabelsMasker.wrapped] loading data from Nifti1Image(
shape=(91, 109, 91),
affine=array([[   2.,    0.,    0.,  -90.],
       [   0.,    2.,    0., -126.],
       [   0.,    0.,    2.,  -72.],
       [   0.,    0.,    0.,    1.]])
)
[Memory]13.3s, 0.2min   : Loading _filter_and_extract...
__________________________________filter_and_extract cache loaded - 0.0s, 0.0min

References

[1]

Gaël Varoquaux and R. Cameron Craddock. Learning and comparing functional connectomes across subjects. NeuroImage, 80:405–415, 2013. Mapping the Connectome. URL: https://www.sciencedirect.com/science/article/pii/S1053811913003340, doi:https://doi.org/10.1016/j.neuroimage.2013.04.007.

[2]

Rastko Ciric, Daniel H. Wolf, Jonathan D. Power, David R. Roalf, Graham L. Baum, Kosha Ruparel, Russell T. Shinohara, Mark A. Elliott, Simon B. Eickhoff, Christos Davatzikos, Ruben C. Gur, Raquel E. Gur, Danielle S. Bassett, and Theodore D. Satterthwaite. Benchmarking of participant-level confound regression strategies for the control of motion artifact in studies of functional connectivity. NeuroImage, 154(1):174–187, 2017. doi:10.1016/j.neuroimage.2017.03.020.

[3]

Linden Parkes, Ben Fulcher, Murat Yücel, and Alex Fornito. An evaluation of the efficacy, reliability, and sensitivity of motion correction strategies for resting-state functional MRI. NeuroImage, 171:415–436, May 2018. doi:10.1016/j.neuroimage.2017.12.073.

Estimated memory usage: 759 MB