fmri_bin

Description

FMRI_BIN Bins continuous brain data into a set of discrete colour bands. Assigns each voxel to one of NBINS discrete bins spanning [MINVAL, MAXVAL]. Returns a binary (228453 x NBINS) matrix where column k is 1 for voxels whose value falls within bin k, and a corresponding RGB colour lookup table suitable for multi-colour 3D rendering. Used internally by fmri_tfce and fmri_show3d for multi-level rendering and cluster-size integration.

Usage

[bindata, bincmap] = fmri_bin(data, minval, maxval)
[bindata, bincmap] = fmri_bin(data, minval, maxval, nbins, cmap)

Inputs

DATA: Statistical map, vectorized (228453 x 1).
MINVAL: Lower bound: values below are excluded.
MAXVAL: Upper bound: values above are clipped to the top bin.
NBINS: Number of discrete bins. Default: 10.
CMAP: (NBINS x 3) RGB colormap. Default: jet, scaled from midrange.

Outputs

BINDATA: Binary matrix (228453 x NBINS). Column k is 1 where data >= bin_k threshold.
BINCMAP: RGB colour matrix (NBINS x 3), one row per bin.

Examples

[bd, bc] = fmri_bin(zmap, 2, 5);
[bd, bc] = fmri_bin(zmap, 2, 5, 20);
fmri_show3d(bd, [], [], [], [], bc)