Meeting Banner
Abstract #2256

Tumor classification and prediction using robust multivariate clustering of multiparametric MRI

Alexis Arnaud 1,2 , Florence Forbes 1,2 , Nicolas Coquery 3,4 , Emmanuel L Barbier 3,4 , and Benjamin Lemasson 3,4

1 INRIA, Grenoble, -, France, 2 LJK, University Grenoble Alpes, Grenoble, -, France, 3 U836, INSERM, Grenoble, -, France, 4 GIN, University Grenoble Alpes, Grenoble, -, France

Multiparametric MRI combined with multidimensional advanced statistical analysis methods may allow a more efficient brain tumor characterization. We used an Expectation-Maximization algorithm and Bayesian model selection on small animal data collected at 4.7T and using four glioma models (n=37). We first detected and excluded outlier animals (n=1). Then, we built a dictionary of tumor signatures. This dictionary discriminated four tumor models.

This abstract and the presentation materials are available to members only; a login is required.

Join Here