We focus on the development of time-efficient high-fidelity numerical models that simulate the electromagnetic fields in volume conductor models. Using these models we solve optimization and inverse problems in bio-electromagnetism with special attention towards the accuracy, time-efficiency and complexity of the problems. Fundamental research is performed in scientific computing with respect to the following topics:

  • Surrogate-based optimization strategies (time-efficiency),
  • Uncertainties and robustness in (stochastic) optimization and inverse problems (accuracy),
  • Dealing with the ill-posedness of the inverse problem (accuracy),
  • Coupled forward models acting on multiple time and space scales (complexity),
  • Smart inverse problems: use of inverse problems together with machine learning techniques in feedback loops for better control (complexity),
  • Etc.

The considered numerical techniques are universal and can be applied onto other fields of research. Progress on the technological side of the applications listed below requires comparable progress in the numerical techniques listed above. Applications are situated within the neurology and oncology biomedical field for diagnostics and/or therapy purposes:

Relevant publications

G. Crevecoeur, H. Hallez, P. Van Hese, Y. D’Asseler, L. Dupré, and R. Van de Walle, "EEG source analysis using space mapping techniques," Journal of Computational and Applied Mathematics, Vol. 215, No. 2, pp. 339-347, 2008.

G. Crevecoeur, H. Hallez, P. Van Hese, Y. D'Asseler, L. Dupré, L. Vandenbossche, and R. Van de Walle, "A hybrid algorithm for solveing the EEG inverse problem from spatio-temporal EEG data," Medical and Biological Engineering and Computing, Vol. 46, No. 8, pp. 767-777, 2008. 


External links


For more information:,