Machine learning and BCI

Summary from only given. A main motivation for multimodal imaging has been the possibility to enhance medical diagnosis[1]. Beyond this original medical motivation the fusion of multiple modalities has created successful interesting research opportunities that have furthered our understanding of the brain and cognition[15]. In BCI recently multimodal fusion concepts have received great attention under the label hybrid BCI[13]. Fusing information has also been a very common practice in the sciences and engineering [17]. Recently a family of novel multimodal data analysis methods have emerged that can extract nonlinear relations between data[1,2,5-10]. They are rooted in the modern machine learning and signal processing techniques that are now available for analysing EEG, for decoding mental states etc[3,11,12,14,16]. The talk will first discuss recent multimodal analysis techniques such as SPoC[5-7]. Furthermore if time permits we will discuss a novel reliable method for estimating the Hurst exponent, a quantity that has recently become popular for describing network properties and is being used for diagnostic purposes[4]. Both nonlinear techniques allow for a better and more reliable and robust analysis of complex phenomena in neurophysiological data.