Skip to main navigationSkip to main content
The University of Southampton
Southampton Neuroscience GroupOur news, seminars & events

Congratulations to Mamun who presented his poster at Westminster as one of the finalists in SET for Britain 2011

Published: 22 March 2011

Congratulations to Mamun who presented his poster at Westminster as one of the finalists in SET for Britain 2011

Recognition of Voluntary Movement from Human Subthalamic Activity for Brain Computer Interface

People with disability due to stroke, quadriplegia have difficulty nteracting with the environment to perform daily tasks. To improve patients' quality of living, feasible assistive technology is needed.

Brain Computer Interface (BCI) is one promising and challenging
approach providing direct communication pathway between brain and machine or robotics. Research on BCI not only provides tools for helping people by assisting, recovering or augmenting human cognitive or sensory-motor functions, but also decoding of brain activity required to assist in the building of Brain-Machine-Brain Interface for treatment of diseases or investigation of neural circuit mechanisms.

The integration of BCI technology into Deep Brain Stimulation (DBS), a unique interface to sense and intervene the human brain circuits, may advance its intelligence and way of evoking brain plasticity. The pathological or physiological states have been investigated in the neural activity of Subthalamic Nucleus (STN), a neural centre involved in Parkinson's disease and motor control.

Local field potentials (LFPs) related to voluntary movement were
recorded from bilateral STNs via implanted deep brain  stimulation electrodes in patients with Parkinson's disease while they performed Left or right finger clicking task with visually guided cues.

The objective of this study is to decode the neural activity related to human voluntary movements from STN LFPs using signal processing and classification algorithms.

Important features of STN LFPs related to movement executions and laterality were extracted by wavelet packet transform and a support vector machine classifier was employed to sequentially recognise the occurrence of movement and whether the forthcoming movement was by left or right hand.

Privacy Settings