Article: "MVPAlab: an intuitive tool for multivariate patterns analysis in magneto / electroencephalography data"
Researchers from the UGR; CIMCYC and DaSCI, have collaborated to develop a tool for the analysis of multivariate patterns in brain activity data obtained using the electroencephalography technique.
MVPAlab, created by Dávid López García, Chema G. Peñalver, Juanma Górriz and María Ruz, is based on Matlab and offers an intuitive and easy-to-use interface that does not require programming skills.
Among the functionalities that MVPAlab offers are the possibility to configure different types of multivariate analysis and to graphically represent the results in a visually attractive way. In addition, MVPAlab implements numerous algorithms based on machine learning that allow different types of statistical analysis to be performed. There are also available a set of subroutines that perform specific tasks such as normalizing, balancing, and reducing the dimensions of the data.
The MVPAlab source code is hosted on a GitHub repository (https://github.com/dlopezg/mvpalab). Users can use, modify and share this tool freely, as well as find different tutorials, or a discussion forum where they can suggest new features.
MVPAlab continues to develop to implement new features in the future. The use of these techniques represents a step forward in the study and understanding of human brain function.
Full reference:
López-García, D., Peñalver, J.M.G., Górriz, J.M. & Ruz, M. (en prensa) MVPAlab: A Machine Learning decoding toolbox for multidimensional electroencephalography data. Computer Methods and Programs in Biomedicine https://www.biorxiv.org/content/10.1101/2021.06.24.449693v2
Researchers contact:
- David López García, dlopez@ugr.es
- Chema G. Peñalver, cgpenalver@ugr.es
- María Ruz, mruz@ugr.es