Note

This page is a reference documentation. It only explains the class signature, and not how to use it. Please refer to the user guide for the big picture.

nilearn.glm.first_level.FirstLevelModel¶

class nilearn.glm.first_level.FirstLevelModel(t_r=None, slice_time_ref=0.0, hrf_model='glover', drift_model='cosine', high_pass=0.01, drift_order=1, fir_delays=None, min_onset=-24, mask_img=None, target_affine=None, target_shape=None, smoothing_fwhm=None, memory=None, memory_level=1, standardize=False, signal_scaling=0, noise_model='ar1', verbose=0, n_jobs=1, minimize_memory=True, subject_label=None, random_state=None)[source]¶

Implement the General Linear Model for single run fMRI data.

Parameters:

t_rfloat or None, default=None

This parameter indicates repetition times of the experimental runs. In seconds. It is necessary to correctly consider times in the design matrix. This parameter is also passed to nilearn.signal.clean. Please see the related documentation for details.

slice_time_reffloat, default=0.0

This parameter indicates the time of the reference slice used in the slice timing preprocessing step of the experimental runs. It is expressed as a fraction of the t_r (repetition time), so it can have values between 0. and 1.

hrf_modelstr, function, list of functions, or None

This parameter defines the HRF model to be used. It can be a string if you are passing the name of a model implemented in Nilearn. Valid names are:

"spm": This is the HRF model used in SPM. See spm_hrf.
"spm + derivative": SPM model plus its time derivative. This gives 2 regressors. See spm_hrf, and spm_time_derivative.
"spm + derivative + dispersion": Same as above plus dispersion derivative. This gives 3 regressors. See spm_hrf, spm_time_derivative, and spm_dispersion_derivative.
"glover": This corresponds to the Glover HRF. See glover_hrf.
"glover + derivative": The Glover HRF + time derivative. This gives 2 regressors. See glover_hrf, and glover_time_derivative.
"glover"+ derivative + dispersion": Same as above plus dispersion derivative. This gives 3 regressors. See glover_hrf, glover_time_derivative, and glover_dispersion_derivative.
"fir": Finite impulse response basis. This is a set of delayed dirac models.

It can also be a custom model. In this case, a function should be provided for each regressor. Each function should behave as the other models implemented within Nilearn. That is, it should take both t_r and oversampling as inputs and return a sample numpy array of appropriate shape.

Note

It is expected that "spm" standard and "glover" models would not yield large differences in most cases.

Note

In case of "glover" and "spm" models, the derived regressors are orthogonalized with respect to the main one.

Default=’glover’.

drift_modelstr, default=’cosine’

This parameter specifies the desired drift model for the design matrices. It can be ‘polynomial’, ‘cosine’ or None.

high_passfloat, default=0.01

This parameter specifies the cut frequency of the high-pass filter in Hz for the design matrices. Used only if drift_model is ‘cosine’.

drift_orderint, default=1

This parameter specifies the order of the drift model (in case it is polynomial) for the design matrices.

fir_delaysarray of shape(n_onsets), list or None, default=None

Will be set to [0] if None is passed. In case of FIR design, yields the array of delays used in the FIR model, in scans.

min_onsetfloat, default=-24

This parameter specifies the minimal onset relative to the design (in seconds). Events that start before (slice_time_ref * t_r + min_onset) are not considered.

mask_imgNiimg-like, NiftiMasker, SurfaceImage, SurfaceMasker, False or None, default=None

Mask to be used on data. If an instance of masker is passed, then its mask will be used. If None is passed, the mask will be computed automatically by a NiftiMasker or SurfaceMasker with default parameters. If False is given then the data will not be masked. In the case of surface analysis, passing None or False will lead to no masking.

target_affine3x3 or 4x4 matrix, or None, default=None

This parameter is passed to nilearn.image.resample_img. Please see the related documentation for details.

target_shape3-tuple of int, or None, default=None

This parameter is passed to nilearn.image.resample_img. Please see the related documentation for details.

smoothing_fwhmfloat, optional.

If smoothing_fwhm is not None, it gives the full-width at half maximum in millimeters of the spatial smoothing to apply to the signal.

memorystr or pathlib.Path, default=None

Path to the directory used to cache the masking process and the glm fit. By default, no caching is done. Creates instance of joblib.Memory. If None is passed will default to Memory(location=None).

memory_levelint or None, default=None

Rough estimator of the amount of memory used by caching. Higher value means more memory for caching.

standardizebool, default=False

If standardize is True, the time-series are centered and normed: their variance is put to 1 in the time dimension.

signal_scalingFalse, int or (int, int), default=0

If not False, fMRI signals are scaled to the mean value of scaling_axis given, which can be 0, 1 or (0, 1). 0 refers to mean scaling each voxel with respect to time, 1 refers to mean scaling each time point with respect to all voxels & (0, 1) refers to scaling with respect to voxels and time, which is known as grand mean scaling. Incompatible with standardize (standardize=False is enforced when signal_scaling is not False).

noise_model{‘ar1’, ‘ols’}, default=’ar1’

The temporal variance model.

verboseint, default=0

Indicate the level of verbosity. By default, nothing is printed. If 0 prints nothing. If 1 prints progress by computation of each run. If 2 prints timing details of masker and GLM. If 3 prints masker computation details.

n_jobsint, default=1

The number of CPUs to use to do the computation. -1 means ‘all CPUs’, -2 ‘all CPUs but one’, and so on.

minimize_memorybool, default=True

Gets rid of some variables on the model fit results that are not necessary for contrast computation and would only be useful for further inspection of model details. This has an important impact on memory consumption.

subject_labelstr, optional

This id will be used to identify a FirstLevelModel when passed to a SecondLevelModel object.

random_stateint or numpy.random.RandomState, default=None.

Random state seed to sklearn.cluster.KMeans for autoregressive models of order at least 2 (‘ar(N)’ with n >= 2).

Added in version 0.9.1.

Attributes:

labels_array of shape (n_voxels,),: a map of values on voxels used to identify the corresponding model
results_dict,: with keys corresponding to the different labels values. Values are SimpleRegressionResults corresponding to the voxels, if minimize_memory is True, RegressionResults if minimize_memory is False

__init__(t_r=None, slice_time_ref=0.0, hrf_model='glover', drift_model='cosine', high_pass=0.01, drift_order=1, fir_delays=None, min_onset=-24, mask_img=None, target_affine=None, target_shape=None, smoothing_fwhm=None, memory=None, memory_level=1, standardize=False, signal_scaling=0, noise_model='ar1', verbose=0, n_jobs=1, minimize_memory=True, subject_label=None, random_state=None)[source]¶

fit(run_imgs, events=None, confounds=None, sample_masks=None, design_matrices=None, bins=100)[source]¶

Fit the GLM.

For each run: 1. create design matrix X 2. do a masker job: fMRI_data -> Y 3. fit regression to (Y, X)

Parameters:

run_imgsNiimg-like object, list or tuple of Niimg-like objects, SurfaceImage object, or list or tuple of SurfaceImage: Data on which the GLM will be fitted. If this is a list, the affine is considered the same for all.

Warning

If the FirstLevelModel object was instantiated with a mask_img, then run_imgs must be compatible with mask_img. For example, if mask_img is a nilearn.maskers.NiftiMasker instance or a Niimng-like object, then run_imgs must be a Niimg-like object, a list or a tuple of Niimg-like objects. If mask_img is a SurfaceMasker or SurfaceImage instance, then run_imgs must be a SurfaceImage, a list or a tuple of SurfaceImage.
eventspandas.DataFrame or str or list of pandas.DataFrame or str, default=None: fMRI events used to build design matrices. One events object expected per run_img. Ignored in case designs is not None. If string, then a path to a csv or tsv file is expected. See make_first_level_design_matrix for details on the required content of events files.
confoundspandas.DataFrame, numpy.ndarray or str or list of pandas.DataFrame, numpy.ndarray or str, default=None: Each column in a DataFrame corresponds to a confound variable to be included in the regression model of the respective run_img. The number of rows must match the number of volumes in the respective run_img. Ignored in case designs is not None. If string, then a path to a csv file is expected.
sample_masksarray_like, or list of array_like, default=None: shape of array: (number of scans - number of volumes remove) Indices of retained volumes. Masks the niimgs along time/fourth dimension to perform scrubbing (remove volumes with high motion) and/or remove non-steady-state volumes.

Added in version 0.9.2.
design_matricespandas.DataFrame or list of pandas.DataFrame, default=None: Design matrices that will be used to fit the GLM. If given it takes precedence over events and confounds.
binsint, default=100: Maximum number of discrete bins for the AR coef histogram. If an autoregressive model with order greater than one is specified then adaptive quantification is performed and the coefficients will be clustered via K-means with bins number of clusters.

compute_contrast(contrast_def, stat_type=None, output_type='z_score')[source]¶

Generate different outputs corresponding to the contrasts provided e.g. z_map, t_map, effects and variance.

In multi-run case, outputs the fixed effects map.

Parameters:

contrast_defstr or array of shape (n_col) or list of (string or array of shape (n_col)): where n_col is the number of columns of the design matrix, (one array per run). If only one array is provided when there are several runs, it will be assumed that the same contrast is desired for all runs. One can use the name of the conditions as they appear in the design matrix of the fitted model combined with operators +- and combined with numbers with operators +-*/. In this case, the string defining the contrasts must be a valid expression for compatibility with pandas.DataFrame.eval.
stat_type{‘t’, ‘F’}, optional: Type of the contrast.
output_typestr, default=’z_score’: Type of the output map. Can be ‘z_score’, ‘stat’, ‘p_value’, ‘effect_size’, ‘effect_variance’ or ‘all’.

Returns:

outputNifti1Image, SurfaceImage, or dict: The desired output image(s). If output_type == 'all', then the output is a dictionary of images, keyed by the type of image.

fit_transform(X, y=None, **fit_params)¶

Fit to data, then transform it.

Fits transformer to X and y with optional parameters fit_params and returns a transformed version of X.

Parameters:

Xarray-like of shape (n_samples, n_features): Input samples.
yarray-like of shape (n_samples,) or (n_samples, n_outputs), default=None: Target values (None for unsupervised transformations).
**fit_paramsdict: Additional fit parameters.

Returns:

X_newndarray array of shape (n_samples, n_features_new): Transformed array.

generate_report(contrasts, title=None, bg_img='MNI152TEMPLATE', threshold=3.09, alpha=0.001, cluster_threshold=0, height_control='fpr', min_distance=8.0, plot_type='slice', display_mode=None, report_dims=(1600, 800))[source]¶

Return a HTMLReport which shows all important aspects of a fitted GLM.

The HTMLReport can be opened in a browser, displayed in a notebook, or saved to disk as a standalone HTML file.

The GLM must be fitted and have the computed design matrix(ces).

Note

The FirstLevelModel or SecondLevelModel must have been fitted prior to calling generate_report.

Parameters:

contrastsdict [ str, ndarray ] or str or list [ str ] or ndarray or list [ ndarray ]

Contrasts information for a FirstLevelModel or SecondLevelModel.

Example:

Dict of contrast names and coefficients, or list of contrast names or list of contrast coefficients or contrast name or contrast coefficient

Each contrast name must be a string. Each contrast coefficient must be a list or numpy array of ints.

Contrasts are passed to contrast_def for FirstLevelModel through compute_contrast, and second_level_contrast for SecondLevelModel through compute_contrast.

titlestr, optional

If a str, it represents the web page’s title and primary heading, model type is sub-heading.
If None, page titles and headings are autogenerated using contrast names.

bg_imgNiimg-like object, default=’MNI152TEMPLATE’

See Input and output: neuroimaging data representation. The background image for mask and stat maps to be plotted on upon. To turn off background image, just pass “bg_img=None”.

thresholdfloat, default=3.09

Cluster forming threshold in same scale as stat_img (either a t-scale or z-scale value). Used only if height_control is None.

alphafloat, default=0.001

Number controlling the thresholding (either a p-value or q-value). Its actual meaning depends on the height_control parameter. This function translates alpha to a z-scale threshold.

cluster_thresholdint, default=0

Cluster size threshold, in voxels.

height_controlstr or None, default=’fpr’

False positive control meaning of cluster forming threshold: ‘fpr’, ‘fdr’, ‘bonferroni’ or None.

min_distancefloat, default=8.0

For display purposes only. Minimum distance between subpeaks in mm.

plot_type{‘slice’, ‘glass’}, default=’slice’

Specifies the type of plot to be drawn for the statistical maps.

display_mode{‘ortho’, ‘x’, ‘y’, ‘z’, ‘xz’, ‘yx’, ‘yz’, ‘l’, ‘r’, ‘lr’, ‘lzr’, ‘lyr’, ‘lzry’, ‘lyrz’}, optional

Choose the direction of the cuts:

‘x’ - sagittal

‘y’ - coronal

‘z’ - axial

‘l’ - sagittal left hemisphere only

‘r’ - sagittal right hemisphere only

‘ortho’ - three cuts are performed in orthogonal directions

Default is ‘z’ if plot_type is ‘slice’; ‘ortho’ if plot_type is ‘glass’.

report_dimsSequence[ int, int ], default=(1600, 800)

Specifies width, height (in pixels) of report window within a notebook. Only applicable when inserting the report into a Jupyter notebook. Can be set after report creation using report.width, report.height.

Returns:

report_textHTMLReport: Contains the HTML code for the GLM report.

get_metadata_routing()¶

Get metadata routing of this object.

Please check User Guide on how the routing mechanism works.

Returns:

routingMetadataRequest: A MetadataRequest encapsulating routing information.

get_params(deep=True)¶

Get parameters for this estimator.

Parameters:

deepbool, default=True: If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns:

paramsdict: Parameter names mapped to their values.

predicted()[source]¶

Transform voxelwise predicted values to the same shape as the input Nifti1Image(s).

Returns:

outputlist: A list of Nifti1Image(s).

r_square()[source]¶

Transform voxelwise r-squared values to the same shape as the input Nifti1Image(s).

Returns:

outputlist: A list of Nifti1Image(s).

residuals()[source]¶

Transform voxelwise residuals to the same shape as the input Nifti1Image(s).

Returns:

outputlist: A list of Nifti1Image(s).

set_fit_request(*, bins='$UNCHANGED$', confounds='$UNCHANGED$', design_matrices='$UNCHANGED$', events='$UNCHANGED$', run_imgs='$UNCHANGED$', sample_masks='$UNCHANGED$')¶

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.
False: metadata is not requested and the meta-estimator will not pass it to fit.
None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.
str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

binsstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED: Metadata routing for bins parameter in fit.
confoundsstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED: Metadata routing for confounds parameter in fit.
design_matricesstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED: Metadata routing for design_matrices parameter in fit.
eventsstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED: Metadata routing for events parameter in fit.
run_imgsstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED: Metadata routing for run_imgs parameter in fit.
sample_masksstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED: Metadata routing for sample_masks parameter in fit.

Returns:

selfobject: The updated object.

set_output(*, transform=None)¶

Set output container.

See Introducing the set_output API for an example on how to use the API.

Parameters:

transform{“default”, “pandas”}, default=None

Configure output of transform and fit_transform.

“default”: Default output format of a transformer
“pandas”: DataFrame output
“polars”: Polars output
None: Transform configuration is unchanged

Added in version 1.4: “polars” option was added.

Returns:

selfestimator instance: Estimator instance.

set_params(**params)¶

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters:

**paramsdict: Estimator parameters.

Returns:

selfestimator instance: Estimator instance.

Examples using `nilearn.glm.first_level.FirstLevelModel`¶

Intro to GLM Analysis: a single-run, single-subject fMRI dataset

Decoding of a dataset after GLM fit for signal extraction

Default Mode Network extraction of ADHD dataset

Analysis of an fMRI dataset with a Finite Impule Response (FIR) model

Example of surface-based first-level analysis

Single-subject data (two runs) in native space

Predicted time series and residuals

First level analysis of a complete BIDS dataset from openneuro

Simple example of two-runs fMRI model fitting

Understanding parameters of the first-level model

BIDS dataset first and second level analysis

Surface-based dataset first and second level analysis of a dataset

Beta-Series Modeling for Task-Based Functional Connectivity and Decoding

nilearn.glm.first_level.FirstLevelModel¶

Examples using nilearn.glm.first_level.FirstLevelModel¶

Examples using `nilearn.glm.first_level.FirstLevelModel`¶