joel carlson
2015-Dec-11 08:44 UTC
[R-pkgs] New Package "radiomics": First and Second Order Matrix Statistics
Dear R Users, I am pleased to inform you that my package, 'radiomics' is now available on CRAN: https://cran.r-project.org/web/packages/radiomics/ <http://cran.r-project.org/web/packages/clifro/> The radiomics package offers 4 classes of texture matrices, and associated feature sets. These matrices can be used for image classification, or for extraction of meaningful information from the input image (as in the case of the research field of radiomics). The texture matrices are: - Gray level co-occurrence matrix (GLCM) - Gray level run-length matrix (GLRLM) - Gray level size-zone matrix (GLSZM) - Multiple gray level size-zone matrix (MGLSZM) Each of these matrices are calculated from an input grayscale matrix, and return objects of class 'glcm', 'glrlm', 'glszm' or 'mglszm'. Features of the matrices can be calculated using the calc_features() function. calc_features() can also be called on an image matrix, and will return first-order summary statistics. The github page can be viewed here:https://github.com/joelcarlson/radiomics <https://github.com/ropensci/clifro#enhancing-the-national-climate-database-with-clifro> Lists of features, and an introduction to the package can be found in the package vignette. In-depth explanation of texture analysis matrices can be found here: http://joelcarlson.me/2015/07/10/radiomics-package/ Any comments, contributions, and feedback are always welcome. I hope you find this helpful! Regards --- Joel Carlson [[alternative HTML version deleted]]