AI for spectral analysis
We are pleased to announce that our MRS team has participated in the ISBI Edited MRS Reconstruction Challenge and achieved great success. Out of three categories, our team is the winner of track 1 and runner-up in tracks 2 and 3.
More info can be found here.
On 19 April 2023 we organized a workshop at TU/e on the generizability of AI in spectroscopy. Apart from the Spectralligence partners, representatives were invited from various institutes and companies active in different fields of spectroscopy (Bruker, Spectro AG, Wageningen University, Jena University, Avantes). Photo: PhD researcher Dennis van de Sande opens the workshop. More info can be found here.
PhD researchers Julian Merkofer (left picture) and Dennis van de Sande (right picture) presented posters at the ISMRM 2023 Benelux conference, that was held on 24 March 2023 in Brussels. Some details can be found here.
On 1 March 2023, we organized a “Transformer hackathon”, during which we discussed if and how the new concept of “Transformer neural networks” can be used to enhance the various spectral analysis applications that we study in our Spectralligence project. Guest lectures were given by dr. Mitko Veta (from partner TU/e) and prof.dr. Marc Aubreville (Ingolstadt Medical Imaging).
Two abstracts about Spectralligence results have been accepted for presentation by the ISMRM 2023 conference:
Julian Merkofer: Predicting Uncertainty of Metabolite Quantification in Magnetic Resonance Spectroscopy with Applications for Adaptive Ensembling.
Dennis van de Sande: Conditional VAE for Single-Voxel MRS Data Generation.
On 1 November 2021 the ITEA4 Spectralligence project has started, and on 10 November we held our first in-person meeting at Philips Healthcare in Best, The Netherlands, with online participation of our Finnish partner.