matrixCreator: Convert Database to Functional Data Matrix Format

In neuroSCC: Bridging Simultaneous Confidence Corridors and PET Neuroimaging

Convert Database to Functional Data Matrix Format

Description

Converts a PET image database (created via databaseCreator) into a matrix format suitable for functional data analysis. Each row of the resulting matrix corresponds to a subject, and each column to a voxel's PET intensity values at a specified brain slice.

Usage

matrixCreator(
  database,
  paramZ = 35,
  useSequentialNumbering = FALSE,
  quiet = FALSE
)

Arguments

database	A data frame created by databaseCreator, containing voxel-level PET image data, including subject identifiers, coordinates, and intensity values.
paramZ	An integer specifying the z-coordinate (slice) to extract. Default is 35.
useSequentialNumbering	logical. If TRUE, assigns sequential subject IDs instead of extracting them from filenames. Not currently used inside this function. Default is FALSE.
quiet	logical. If TRUE, suppresses progress messages. Default is FALSE.

Details

This function performs the following steps

1. Verifies that the specified z-slice exists in the database.

2. Identifies the correct subject grouping column (CN_number or AD_number).

3. Determines the matrix dimensions using x and y coordinates.

4. Extracts PET intensities per subject at the given slice.

5. Replaces any NaN values with 0 to ensure numerical stability.

This function typically follows databaseCreator and precedes meanNormalization in the neuroSCC workflow.

Value

A numeric matrix where

• Each row represents one subject's PET values at the selected z-slice.

• Each column corresponds to a voxel (flattened as a 1D row).

See Also

databaseCreator for generating the input database.
meanNormalization for scaling matrix data prior to SCC computation.

Examples

# NOTE: To keep example runtime short, only one synthetic PET file is used.
# For full analysis, expand the filename pattern accordingly.

# Step 1: Generate a database for a single subject
controlPattern <- "^syntheticControl1\\.nii\\.gz$"
databaseControls <- databaseCreator(pattern = controlPattern, control = TRUE, quiet = TRUE)

# Step 2: Convert the database into a matrix format
matrixControls <- matrixCreator(databaseControls, paramZ = 35, quiet = TRUE)

# Display dimensions of the matrix
dim(matrixControls)

neuroSCC documentation built on April 4, 2025, 3:50 a.m.