John Muschelli

Assistant Scientist, Department of Biostatistics

PhD in Biostatistics

• worked on fMRI, sMRI, and CT for ≈ 7 years

Email: [email protected]

Kristin Linn

Assistant Professor, Department of Biostatistics and Epidemiology Perelman School of Medicine, University of Pennsylvania

PhD in Statistics

• worked on sMRI for 3 years

Email: [email protected]

• Multiple sclerosis (MS) is a chronic disease of the central nervous system (brain, spinal chord, optic nerves)
• MS lesions in the brain are areas of active inflammation, demylenation, or permanent tissue damage.
  • lesions are primarily in the white matter
• The full data is available at https://smart-stats-tools.org.

• MRI is well-suited for assessing lesion burden (volume and patterns) because lesions appear as hyperintensities on FLAIR and T2-w images and as hypointensities on T1-w images.
• Using 5 training and 3 test subjects data from the an open MS data set (http://lit.fe.uni-lj.si/tools.php?lang=eng) (Lesjak et al. 2017)

There are multiple imaging formats. We will use NIfTI:

• Neuroimaging Informatics Technology Initiative (https://nifti.nimh.nih.gov/nifti-1)
  • essentially a header and data (binary format)
  • will have extension .nii (uncompressed) or .nii.gz (compressed)
  • we will use 3-D images (4-D and 5-D are possible)
• ANALYZE 7.5 was a precursor to NIfTI
  • had a hdr file (header) and img file (data)

The Course overview is located at (with slides): http://johnmuschelli.com/ISBI_2017.html

All materials for this course (including source for the slides) is located at:

https://github.com/muschellij2/imaging_in_r

RStudio Server

For this course, we will use an RStudio Server because installing all the packages can be a lengthy process. Mostly all the code we show requires a Linux/Mac OSX platform for FSL and other systems:

http://johnmuschelli.com/rstudio_server.html

Virtual Machine

If you want to run things locally, we have a Virtual Machine you can download and install: https://neuroconductor.org/neuroc-vm

• R is a language and environment for statistical computing and graphics

• R was implemented over the S language, which was developed by Bell laboratories

• R is both open source and open development

(source: http://www.r-project.org/)

• Powerful and flexible

• Free (open source)

• Extensive add-on software (packages)

• Designed for statistical computing

• High level language

Why not R?

• Fairly steep learning curve

• Little centralized support, relies on online community and package developers

• Slower, and more memory intensive, than the more traditional programming languages (C, Java, Perl, Python)

• R revolves around functions
  • Commands that take input, performs computations, and returns results
  • When you download R, it has a “base” set of functions/packages (base R)
• Functions are enclosed in packages
  • A package is collection of functions, documentation, data, and tutorials (called vignettes).
  • Written by R users/developers (like us) - some are bad
  • You install a package using the install.packages command/function:

install.packages("oro.nifti")

install.packages is a function, "oro.nifti" is a character string.

• Most packages are in a repository Comprehensive R Archive Network (CRAN)
• Neuroconductor (https://neuroconductor.org/) has packages related to medical imaging

source("https://neuroconductor.org/neurocLite.R")
neuro_install("oro.nifti")

When you install a package, it’s downloaded to the hard disk. That doesn’t mean that you can use the functions from that package just yet.

• You “load”/import a package into memory using the library command

For example, to load the oro.nifti package:

library(oro.nifti)

Now, functions from the oro.nifti package can be used.

RStudio is an Integrated Development Environment (IDE) for R (made by RStudio the company)

• It helps the user effectively and more easily use R.
  • Syntax highlighting, code completion, and smart indentation
  • Easily manage multiple working directories and projects
• Is NOT dropdown statistical tools (such as Stata)
  • See Rcmdr or Radiant

• Where code is executed (where things happen)
• You can type here for things interactively
• Code is not saved on your disk
  • Can act as a calculator
  • Creating variables/objects
  • Applying functions

• Where files open to
• Have R code and comments in them
  • Static copy of what you did (reproducibility)
  • Try things out interactively, then add to your script
• Can highlight and press (CMD+Enter (Mac) or Ctrl+Enter (Windows)) to run the code
• Code is saved to disk

• Tells you what objects are in R
• What exists in memory/what is loaded?/what did I read in?

History

• Shows previous commands. Good to look at for debugging, but don’t rely on it as a script. Make a script!
• Also type the “up” key in the Console to scroll through previous commands

• Files - shows the files on your computer of the directory you are working in
• Viewer - can view data or R objects
• Help - shows help of R commands
• Plots - pretty pictures
• Packages - list of R packages that are loaded in memory

http://johnmuschelli.com/imaging_in_r