nipype.algorithms.icc module¶
ICC¶
Bases: BaseInterface
Calculates Interclass Correlation Coefficient (3,1) as defined in P. E. Shrout & Joseph L. Fleiss (1979). “Intraclass Correlations: Uses in Assessing Rater Reliability”. Psychological Bulletin 86 (2): 420-428. This particular implementation is aimed at relaibility (test-retest) studies.
- Mandatory Inputs:
mask (a pathlike object or string representing an existing file)
subjects_sessions (a list of items which are a list of items which are a pathlike object or string representing an existing file) – N subjects m sessions 3D stat files.
- Outputs:
icc_map (a pathlike object or string representing an existing file)
session_var_map (a pathlike object or string representing an existing file) – Variance between sessions.
subject_var_map (a pathlike object or string representing an existing file) – Variance between subjects.
- nipype.algorithms.icc.ICC_projection_matrix(shape)¶
- nipype.algorithms.icc.ICC_rep_anova(Y, projection_matrix=None)¶
the data Y are entered as a ‘table’ ie subjects are in rows and repeated measures in columns
One Sample Repeated measure ANOVA
Y = XB + E with X = [FaTor / Subjects]
ICC_rep_anova
involves an expensive operation to compute a projection matrix, which depends only on the shape ofY
, which is computed by callingICC_projection_matrix(Y.shape)
. If arrays of multiple shapes are expected, it may be worth pre-computing and passing directly as an argument toICC_rep_anova
.If only one
Y.shape
will occur, you do not need to explicitly handle these, as the most recently calculated matrix is cached automatically. For example, if you are running the same computation on every voxel of an image, you will see significant speedups.If a
Y
is passed with a new shape, a new matrix will be calculated automatically.