This project will develop methods to analyse complex multivariate time-delay relationships between neurological signals\u00a0recorded from human brains. The analysis will extend Empirical Mode\u00a0Decomposition (EMD) which, although effective, suffers from limitations\u00a0rendering it impractical for the online analysis of the interactions of\u00a0brain signals. These drawbacks will be addressed by building a\u00a0mathematical framework of EMD which will enable to understand the\u00a0behaviour of this method better and will also lead to more robust\u00a0methods of extracting relationships from multivariate brain signals. The\u00a0developed framework will enable reformulation of the algorithm in a way\u00a0that lends itself to fast and efficient implementation on affordable\u00a0parallel computing architectures such as Graphics Processor Units (GPUs)\u00a0leading to a fast parallel EMD analysis opening up this signal\u00a0processing technology to new applications such as Brain Computer Interfaces.<\/p>\n","protected":false},"excerpt":{"rendered":"