Multi-way modelling of high-dimensionality electroencephalographic data

The aim of this study is to investigate whether useful information can be extracted from an electroencephalographic (EEG) data set with a very high number of modes, and to determine which model is the most appropriate for this purpose. The data was acquired during the testing phase of a new drug expected to have effect on the brain activity. The implemented test program (several patients followed in time, different doses, conditions, etc.…) led to a six-way data set. After it was confirmed that the exploratory analysis of this data set could not be handled with classical principal component analysis (PCA), and it was verified that multidimensional structure was present, multi-way methods were used to model the data. It appeared that Tucker 3 was the most suited model. It was possible to extract useful information from this high-dimensionality data. Non-relevant sources of variance (outlying patients for instance) were identified so that they can be removed before the in-depth physiological study is performed.