Parcel-level tSNR across runs (ClusteredNeuroVec)

Computes parcel-wise temporal signal-to-noise ratio across ClusteredNeuroVec runs by aggregating means and variances over time.

Usage

compute_tsnr_parcel(cnv_list)

Arguments

cnv_list

List of neuroim2::ClusteredNeuroVec runs.

Value

Numeric vector of length P with parcel tSNR values.

Details

For each parcel, the mean and variance are computed over concatenated time points across all runs. tSNR is calculated as mean / sd, with both numerator and denominator lower-bounded at 1e-6 to avoid division by zero or degenerate weights. This is the parcel-level analogue of compute_tsnr() for voxel data.

See also

compute_tsnr() for voxel-level tSNR computation

Other spatial metrics: build_spatial_metric(), build_spatial_metric_parcel(), compute_tsnr(), make_laplacian(), make_parcel_laplacian()

Examples

# \donttest{
# Compute parcel tSNR across multiple runs
tsnr_p <- compute_tsnr_parcel(cnv_list)
#> Error: object 'cnv_list' not found
# }