Spiking neural network methodology for modelling, classification and understanding of EEG spatio-temporal data measuring cognitive processes

The paper offers a new methodology for modelling, recognition and understanding of electroencephalography (EEG) spatio-temporal data measuring complex cognitive brain processes during mental tasks. The key element is that mental tasks are performed through complex spatio-temporal brain processes and they can be better understood only if we model properly the spatio-/spectro temporal data that measures these processes. The proposed methodology is based on a recently proposed novel spiking neural network architecture, called NeuCube as a general framework for spatio-temporal brain data modelling. The methodology is demonstrated on benchmark cognitive EEG data. The new approach leads to a faster data processing, improved accuracy of the EEG data classification and improved understanding of this data and the cognitive processes that generated it. The paper concluded that the new methodology is worth exploring further on other spatio-temporal data, measuring complex cognitive brain processes, aiming at using this method for the development of the next generation of brain-computer interfaces and systems for early diagnosis of degenerative brain disease, such as Alzheimer's Disease (AD), and for personalised neuro-rehabilitation systems.