FMradio

The R-package FMradio supports stable prediction and classification with radiomics data through factor-analytic modeling. This support can be invoked irrespective of the imaging modality (such as, e.g., MRI, PET, CT) used to produce the radiomics data.

Installation

If you wish to install the latest version of FMradio directly from the master branch here at GitHub, run

#install.packages("devtools")  # Uncomment if devtools is not installed
devtools::install_github("CFWP/FMradio")

Be sure that you have the package development prerequisites if you wish to install the package from the source.

Manual and other documentation

• A pdf-version of the manual can be found here.

• The R-script used to produce the simulations and results contained in Peeters, C.F.W., et al. (2019) [see references below] can be found here.

References

Relevant publications to FMradio include:

1. Peeters, C.F.W. (2019). "FMradio: Factor modeling for radiomic data". R package, version 1.1.1
2. Peeters, C.F.W., et al. (2019) "Stable prediction with radiomics data". arXiv:1903.11696 [stat.ML]