fmriqa

Overview

The fmriqa package provides an implementation of the fMRI quality assurance (QA) analysis protocol detailed by Friedman and Glover (2006) doi:10.1002/jmri.20583.

Installation

You can install the stable version of fmriqa from CRAN:

install.packages("fmriqa", dependencies = TRUE)

Or the the development version from GitHub (requires devtools package):

install.packages("devtools")
devtools::install_github("martin3141/fmriqa")

Usage

# load the package
library(fmriqa)

# get help on the options for run_fmriqa
?run_fmriqa

# run the analysis - a file chooser will appear when a data_file argument is not given
run_fmriqa()

Real data example

library(fmriqa)
fname <- system.file("extdata", "qa_data.nii.gz", package = "fmriqa")
res <- run_fmriqa(data_file = fname, gen_png = FALSE, gen_res_csv = FALSE,
                  tr = 3, bg_shrink = 15)

## Reading data  : /home/martin/R/x86_64-pc-linux-gnu-library/4.5/fmriqa/extdata/qa_data.nii.gz
## 
## Basic analysis parameters
## -------------------------
## X,Y matrix     : 80x80
## Slices         : 1
## X,Y,Z pix dims : 1x1x1mm
## TR             : 3s
## Slice #        : 1
## ROI width      : 21
## Total vols     : 200
## Analysis vols  : 198
## 
## QA metrics
## ----------
## Mean signal   : 2561.3
## STD           : 6.4
## Percent fluc  : 0.25
## Drift         : 2.06
## Drift fit     : 1.42
## SNR           : 141.3
## SFNR          : 137
## RDC           : 2.84
## TC outlier    : 2.66
## Spec outlier  : 5.12
## MBG percent   : 1.14

Simulation example

library(fmriqa)
library(oro.nifti)

## oro.nifti 0.11.4

# generate random data
set.seed(1)
sim_data <- array(rnorm(80 * 80 * 1 * 100), dim = c(80, 80, 1, 100))
sim_data[20:60, 20:60, 1, ] <- sim_data[20:60, 20:60, 1, ] + 50
sim_nifti <- oro.nifti::as.nifti(sim_data)
fname <- tempfile()
writeNIfTI(sim_nifti, fname)

## [1] "/tmp/Rtmp13OY0H/file5a94514bf0bf.nii.gz"

# perform qa
res <- run_fmriqa(fname, gen_png = FALSE, gen_res_csv = FALSE, t1_canny = 1,
                  t2_canny = 2)

## Reading data  : /tmp/Rtmp13OY0H/file5a94514bf0bf
## 
## Basic analysis parameters
## -------------------------
## X,Y matrix     : 80x80
## Slices         : 1
## X,Y,Z pix dims : 1x1x1mm
## TR             : 1s
## Slice #        : 1
## ROI width      : 21
## Total vols     : 100
## Analysis vols  : 98
## 
## QA metrics
## ----------
## Mean signal   : 50
## STD           : 0.05
## Percent fluc  : 0.09
## Drift         : 0.45
## Drift fit     : 0.06
## SNR           : 51.7
## SFNR          : 51.2
## RDC           : 21.77
## TC outlier    : 2.74
## Spec outlier  : 4.04
## MBG percent   : 0.07

res$snr

## [1] 51.74954

Plot output from real data showing RF spiking artifact