Table Of Contents

Previous topic

modalities.fmri.fmri

Next topic

modalities.fmri.fmristat.invert

This Page

modalities.fmri.fmristat.delay

Module: modalities.fmri.fmristat.delay

Inheritance diagram for nipy.modalities.fmri.fmristat.delay:

This module defines a class to output estimates of delays and contrasts of delays.

Liao, C.H., Worsley, K.J., Poline, J-B., Aston, J.A.D., Duncan, G.H., Evans, A.C. (2002). ‘Estimating the delay of the response in fMRI data.’ NeuroImage, 16:593-606.

Classes

DelayContrast

class nipy.modalities.fmri.fmristat.delay.DelayContrast(fns, weights, formula, IRF=None, name='', rownames=[])

Bases: nipy.fixes.scipy.stats.models.contrast.Contrast

Specify a delay contrast.

Delay contrasts are specified by a sequence of functions and weights, the functions should NOT already be convolved with any HRF. They will be convolved with self.IRF which is expected to be a filter with a canonical HRF and its derivative – defaults to the Glover model.

Weights should have the same number of columns as len(fns), with each row specifying a different contrast.

__init__(fns, weights, formula, IRF=None, name='', rownames=[])
Parameters:
fns : TODO

TODO

weights : TODO

TODO

formula : TODO

TODO

IRF : TODO

TODO

name : string

TODO

rownames : [string]

TODO

compute_matrix(time=None)
Parameters:
time : TODO

TODO

Returns:

None

extract(results)
Parameters:
results : TODO

TODO

Returns:

ContrastResults

isestimable(t)

To estimate the delay, it is assumed that the response contains

(f ** HRF)(t + delta)

for each delay model time series ‘f’. More specifically, it is assumed that

f(t + delta) = c1 * (f ** HRF)(t) + delta * c2 * (f ** dHRF)(t)

where HRF and dHRF are the HRFs for this delay contrast.

This function checks to ensure that the columns

[(f ** HRF)(t), (f ** dHRF(t))]

are in the column space of the fMRI regression model.

Parameters:
t : TODO

TODO

Returns:

None

Raises valueerror:
 

if any of the columns are not in the column space of the model

DelayContrastOutput

class nipy.modalities.fmri.fmristat.delay.DelayContrastOutput(coordmap, contrast, IRF=None, dt=0.01, delta=None, subpath='delays', clobber=False, path='.', ext='.hdr', volume_start_times=[], **kw)

Bases: nipy.algorithms.statistics.regression.TOutput

TODO

__init__(coordmap, contrast, IRF=None, dt=0.01, delta=None, subpath='delays', clobber=False, path='.', ext='.hdr', volume_start_times=[], **kw)
Parameters:
coordmap : TODO

TODO

contrast : TODO

TODO

IRF : TODO

TODO

dt : float

TODO

delta : TODO

TODO

subpath : string

TODO

clobber : bool

TODO

path : string

TODO

ext : string

TODO

volume_start_times : TODO

TODO

kw : dict

Passed through to the constructor of TContrastOutput

extract(results)
Parameters:
results : TODO

TODO

Returns:

TODO

set_next(data)
Parameters:
data : TODO

TODO

Returns:

None

DelayHRF

class nipy.modalities.fmri.fmristat.delay.DelayHRF(input_hrf=<nipy.modalities.fmri.filters.Filter object at 0xb3957cc>, spectral=True, **keywords)

Bases: nipy.modalities.fmri.hrf.SpectralHRF

Delay filter with spectral or Taylor series decomposition for estimating delays.

Liao et al. (2002).

__init__(input_hrf=<nipy.modalities.fmri.filters.Filter object at 0xb3957cc>, spectral=True, **keywords)
Parameters:
input_hrf : TODO

TODO

spectral : bool

TODO

keywords : dict

Passed through as keywords to the hrf.SpectralHRF constructor.

deltaPCA(tmax=50.0, lower=-15.0, delta=None)

Perform an expansion of fn, shifted over the values in delta. Effectively, a Taylor series approximation to fn(t+delta), in delta, with basis given by the filter elements. If fn is None, it assumes fn=IRF[0], that is the first filter.

Parameters:
tmax : float

TODO

lower : float

TODO

delta : [float]

TODO

Returns:

None