Travel Grant: Petra van Houdt (Montreal)
- Published Date18 July 2011
The main topic of our project is the optimization and evaluation of co-registered EEG and functional MRI in patients who are candidates for epilepsy surgery. This project is a collaboration between Kempenhaeghe (dr. Pauly Ossenblok and Prof. dr. Paul Boon), VU medical center (dr. Jan de Munck and Prof. Dr. Kees Stam) and UMC Utrecht (dr. Frans Leijten). EEG-fMRI is a relatively new technique, which has been used as a research tool, but our results and the results of other groups indicate that EEG-fMRI is well suitable as a clinical tool
From the 20th to the 24th of June, I visited the laboratory of Prof. Dr. Jean Gotman from the Montreal Neurological Institute (MNI). Jean Gotman and his group have much experience with EEG-fMRI in patients with epilepsy (http://apps.mni.mcgill.ca/research/gotman/eeg-fmri-main.htm). The aim of my visit was to discuss and share our experiences about EEG-fMRI, such that we might learn some new (practical) tips for the next series of measurements we are planning to acquire in Kempenhaeghe.
Fig. 1 Campus of the McGillUniversity
At the MNI, I gave a presentation about our data, methods, and results so far. Furthermore, I was present during two EEG-fMRI recordings. This led to interesting discussions about the differences between our institutions and the future of EEG-fMRI. In our discussions, we noticed that the quality of the EEG data is the largest limitation of EEG-fMRI. Since the EEG is measured in the MR environment, two types of artefacts occur in the data: image artifacts (i.e. related to the switching of the gradients) and ballistocardiographic artifacts (i.e. related to the heart beat of the patients). The ballistocardiographic artifacts appear to be the most difficult to remove completely due to the inter- and intra-subject variability. At Kempenhaeghe and MNI different methods are used to remove these artifacts. The method used at the MNI is based on independent component analysis (ICA) and our method is based on the subtraction of an average artefact template. Fortunately, it was possible to import our data in their software program. So far, we analysed two data sets and it appeared that for one data set the ICA approach yielded a better result, whereas in the other dataset the subtraction method might have been better. However, this needs to be assessed in more detail. In the future, this topic might be worthwhile to investigate further.
After my visit to the MNI, I went to the Human Brain Mapping Conference in Quebec, which was also very interesting and led to nice discussions as well. I think the combination of the visit to the MNI and the conference, was very fruitful and I learned a lot! I would like to thank the people at the MNI, especially Prof. Jean Gotman, FrancescaPittau, Firas Fahoum, and Renaud Lopes, for their hospitality and interesting discussions. And I would like to thank the Vereniging voor Biofysica en Biomedische Technologie to make this visit possible.