Abstract #3863

MRSI-based characterization of GBM using a novel map: Expected Distance to Tumor (EDT)

Nuno Pedrosa de Barros¹, Raphael Meier², Samuel Stettler¹, Urspeter Knecht¹, Evelyn Herrmann³, Philippe Schucht⁴, Mauricio Reyes², Jan Gralla¹, Roland Wiest¹, and Johannes Slotboom¹

¹Institute for Diagnostic and Interventional Neuroradiology, University of Bern, Bern, Switzerland, ²Institute for Surgical Technology and Biomechanics, University of Bern, Bern, Switzerland, ³Institute of Radiooncology, University of Bern, Bern, Switzerland, ⁴Neurosurgery, University of Bern, Bern, Switzerland

MRSI can detect regions of brain tumor infiltration beyond the tumor borders visible in structural-MRI (sMRI). However, this is often achieved using only a small fraction of the information provided by MRSI, namely Cho/NAA maps only. Here, we present a new machine-learning-based approach that translates the multidimensional information provided by each spectrum into a single measure: the Expected Distance to solid Tumor volume visible in sMRI. The results show that peritumoral spectra carry information on the distance to solid tumor and that EDT maps may improve the characterization of peritumoral tissue changes invisible with structural MRI.

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

Join Here