Electrophysiology Predictive Models

Deep brain stimulation of the subthalamic nucleus (STN) has been shown to benefit individuals living with Parkinson disease. The STN is a very small surgical target (7.0mm x 4.0mm), which makes targeting the nucleus very difficult. Furthermore, the sensorimotor subdivision of the STN is the optimal target location for optimal clinical motor symptom improvement. This subdivision is a much smaller region located in the dorso-lateral aspect, recent studies have show characteristic signal features in this area.

The figure above shows an example feature, Teager energy, extracted from the raw brain recording data. The depth values indicate distance from the surgically planned target (negative values above, positive values below), there are 5 electrodes used during the recording. This figure illustrates that as the electrodes enter the STN, the Teager energy increases above the avergae (dotted line) and indidates that the electrodes have entered a nucleus. Using various features, extracted from the raw signal, this project aims to build a predictive model that informs the surgeon of the most optimal surgical implantation site.

The ability to train a predictive model that is capable of estimating the dorsal/ventral borders of the STN, as well as the sensorimotor aspect, would reduce surgery time and improve patient outcome.