Intellectual Merit: Deep brain stimulation (DBS) is a highly promising therapy for Parkinson's disease (PD). Yet most patients do not get full therapeutic benefit from DBS due to its critical dependence on electrode location, a "sweet spot" in the dorsolateral posterior sensorimotor subunit of the sub-thalamic nucleus (STN), for therapeutic efficacy. PI Cheng was trained at a center where 70% of DBS patients obtained full therapeutic benefit, improving so markedly that they no longer require any PD medications. Such efficacy is atypical even in academic centers because DBS electrode placement is not standardized, scientific, or systematic. We propose to construct a neural modeling, estimation and control framework for STN, which will enable the development of a new surgical tool that will standardize DBS placement: an automated intraoperative closed-loop DBS localization system. Development of this transformative technology requires: 1) neurophysiologic characterization of the "sweet spot". In PD patients, microelectrode recordings will measure single unit spiking activity (action potentials) of STN neurons at different distances from the "sweet spot" and from within it. Point process models will be estimated from this data and will capture complex stochastic relationships between extrinsic (e.g. behavior) and intrinsic (local neural network activity) factors and STN spiking activity. Principled inferential methods will confirm the "sweet spot's" existence and characterize its electrophysiological properties;and computational conductance-based modeling will elucidate the ionic mechanisms underlying the "sweet spot's" physiology. 2) construction of neural estimation and control algorithms for STN DBS. Signal processing and control will derive a robust feature set from STN spiking activity which will reliably predict where the electrode is and will then guide the electrode to the sweet spot. This transformative project requires collaborations between physicians, scientists, mathematicians and engineers with expertise in neurosurgery, neurophysiology, neural signal processing, estimation and modeling, and control theory. For these reasons automation of DBS localization remains largely untapped, giving us the opportunity to lead the scientific development of this next-generation technology. Broader Impact: Due to cost, less than 10% of PD patients worldwide receive DBS. Automating surgical implantation and obviating complex postoperative DBS programming from suboptimal electrode placement would decrease cost, and thus increase patient access. Even greater societal impact, however, would come from improved DBS efficacy, which is life-changing for PD patients. DBS patients of Dr. Cheng have stated that they have been returned to their pre-PD status, and that not just their lives but also the lives of their family members, so long held hostage by a debilitating chronic disease process, have been returned to them. Our proposal attempts to extrapolate these benefits to the larger PD population. Even more importantly, DBS is a nascent procedure holding great promise for many future neurological and psychiatric indications. A technology that improves DBS targeting fidelity and efficacy would hold the potential to improve the lives of millions of patients and their families worldwide. This project will be integrated into curricula in the home and affiliated departments of the PIs. Coursework for signal processing and neuronal spike modeling in the senior undergraduate and graduate levels will gain from our proposal. A graduate level modern control theory course with applications to neural systems will also be developed and offered. Traditional courses in neuroanatomy and neurophysiology will be enhanced by our proposal's insights into the relationships between physiology, anatomy, and function. The PIs also plan to reach out to the academic community by providing representative samples of rare neurophysiological data and analysis code. When cultivated, such a database will provide a platform for investigators around the world to benchmark software algorithms, optimize analog and digital components for new hardware platforms that will process neural signals, and develop a more complete understanding of the mechanisms of DBS. PI Cheng has strong relationships with industry companies including Medtronic, the manufacturer of DBS hardware. We will leverage this to expedite the development and testing of our concept. Our project's outcome may thus have a substantial impact on how DBS systems are designed.

National Institute of Health (NIH)
National Institute of Neurological Disorders and Stroke (NINDS)
Research Project (R01)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1-IFCN-B (51))
Program Officer
Liu, Yuan
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Boston University
Biostatistics & Other Math Sci
Schools of Arts and Sciences
United States
Zip Code
Meng, Liang; Kramer, Mark A; Middleton, Steven J et al. (2014) A unified approach to linking experimental, statistical and computational analysis of spike train data. PLoS One 9:e85269
Gerhard, Felipe; Kispersky, Tilman; Gutierrez, Gabrielle J et al. (2013) Successful reconstruction of a physiological circuit with known connectivity from spiking activity alone. PLoS Comput Biol 9:e1003138
Lepage, Kyle Q; Gregoriou, Georgia G; Kramer, Mark A et al. (2013) A procedure for testing across-condition rhythmic spike-field association change. J Neurosci Methods 213:43-62
Agarwal, Rahul; Sarma, Sridevi V (2012) The effects of DBS patterns on basal ganglia activity and thalamic relay : a computational study. J Comput Neurosci 33:151-67
Lepage, Kyle Q; Macdonald, Christopher J; Eichenbaum, Howard et al. (2012) The statistical analysis of partially confounded covariates important to neural spiking. J Neurosci Methods 205:295-304