Meeting Banner
Abstract #4284

Non-Negative Matrix Factorization for Differentiation of Brain Metastasis and Glioblastoma Multiforme, and Visualization of Tumor Infiltration

Jan Luts1, Teresa Laudadio, 1,2, M. Carmen Martinez-Bisbal3,4, Sofie Van Cauter1, Enrique Molla5, Jose Piquer5, Johan Suykens1, Uwe Himmelreich1, Bernardo Celda3,4, Sabine Van Huffel1

1Katholieke Universiteit Leuven, Leuven, Belgium; 2Istituto Applicazioni Calcolo, CNR, Bari, Italy; 3University of Valencia, Valencia, Spain; 4Networking Research Center on Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN), Valencia, Spain; 5Hospital de La Ribera, Valencia, Spain


This study focuses on the differentiation between solitary brain metastasis and glioblastoma multiforme based on conventional magnetic resonance imaging (MRI) and long TE two-dimensional turbo spectroscopic imaging (2D-TSI) data. Fifteen patients with a brain tumor, nine affected by glioblastoma multiforme and six by metastasis, were considered. Non-negative matrix factorization (NNMF) results in a clear separation of glioblastomas and metastases. The methods allows visualizing the abundances of the normal tissue component, which indicate tumor infiltration. In conclusion, automated processing with NNMF of 2D-TSI enables to visualize metabolic differences between glioblastomas and metastases and enables to visualize tumor infiltration.