Biomedical Signal Processing
This project focuses on developing a robust pattern recognition system to recognize and classify human’s behavior (e.g. language and motor tasks) from different brain electrical signals. The objective of this project is to advance research concerning next generation of closed-loop deep brain stimulation (DBS) systems, which are widely used to treat movement disorders such as Parkinson’s disease (PD). This project explores a unique data set of simultaneous recordings of cortical and subcortical electrical potentials in the human brain (e.g., local field potential) obtained during surgical implantation of a DBS system. PD is a neurodegenerative condition and movement disorder diagnosed on the basis of clinical history and motor signs of tremor, rigidity and bradykinesia.
PD incidence increases with advancing age and peaks among people in their 60s and 70s. DBS is an advanced FDA-approved therapeutic technique for alleviating the PD symptoms especially for whom drug therapy is no longer efficient. DBS of the subthalamic nucleus (STN-DBS) improves motor signs of PD and permits reduction of dopaminergic medication. Existing DBS therapy is open-loop, providing a time invariant stimulation pulse train that is not customized to a patient’s current behavioral task goals. By customizing DBS therapy to a patient’s task using machine learning and signal processing methods, these side effects of stimulation may arise only when they are non-detrimental to the patient’s current goals. This could therefore allow more aggressive DBS parameters (e.g. closed-loop DBS) to be used with a reduced risk for therapy limiting side effects.
