nn_impute: Imputes missing values in a 2D matrix based on the nearest...

In ggbrain: Create Images of Volumetric Brain Data in NIfTI Format Using 'ggplot2' Syntax

Imputes missing values in a 2D matrix based on the nearest non-missing neighbors in a given radius

Description

Imputes missing values in a 2D matrix based on the nearest non-missing neighbors in a given radius

Arguments

in_mat	a 2D matrix to fill using nearest neighbors
neighbors	the number of closest non-NA neighboring values to return within in_mat. Default is 4.
radius	the radius (in pixels) around each missing value to search for non-missing neighbors. Default is 8.
aggfun	the function used to aggregate the neighbors in imputation. Supports "mean", "median", and "mode."
ignore_zeros	if TRUE, then zero is not a valid imputation value (since these are not data in NIfTIs)

Details

The "mode" aggfun should only be used when the matrix in_mat can be converted to integers without loss of information (i.e., the data are integerish values already).

Value

A copy of the matrix with NA values imputed by their nearest neighbors

ggbrain documentation built on April 4, 2025, 5:03 a.m.