fmri_timeseries methods, organized by area

fmri_timeseries is a subclass of fmri_data (and therefore of image_vector) specialized for 4-D time series data. It inherits all fmri_data and image_vector properties and methods (data storage in .dat as [voxels x images], .volInfo, masks, history, etc.) and adds a few timeseries-specific fields: a TR, an embedded fmri_glm_design_matrix for first-level GLM specification, and a processing-status table that tracks slice-timing, realignment, denoising, filtering, normalization, and smoothing state.

Because the class inherits from fmri_data, the vast majority of methods relevant to a timeseries (preprocess, regress, predict, ica, mahal, montage, plot, write, etc.) live on the parent classes; see fmri_data methods. The table below lists only methods defined on @fmri_timeseries itself. Type methods(my_obj) in MATLAB for the live list on any instance.

Properties

Property	Description
glm_design_obj	Embedded fmri_glm_design_matrix for first-level model specification
processing_status_table	Table tracking slice-timing, realignment, denoising, filter cutoffs, normalization template, smoothing FWHM
TR	Repetition time in seconds

Inherited from fmri_data / image_vector: dat, volInfo, mask, fullpath, image_names, removed_images, removed_voxels, history, X, Y, covariates, images_per_session, additional_info, etc.

Methods specific to @fmri_timeseries

Method	From	One-liner
fmri_timeseries	@fmri_timeseries	Constructor: load images with a mask, attach TR and a default GLM design
filloutliers	@fmri_timeseries	Per-voxel moving-median outlier replacement with spline interpolation (60-s window)

For everything else (resampling, preprocessing, regression, prediction, extraction, display, I/O), see the inherited methods documented in fmri_data methods.