Date of Thesis

Spring 2022



Neuromodulation is the altering of neural activity using external mechanisms, including but not limited to electrical stimulation and chemical agents. Deep brain stimulation (DBS) is a type of neuromodulation that is largely used to treat movement disorders, such as Parkinson disease (PD) and essential tremor. Recent DBS research has also showed promising results in neuropsychological disorders, such as treatment-resistant depression. The DBS treatment procedure involves implanting electrodes in a region of the brain associated with a particular condition that is producing abnormal signals to apply electrical pulses which essentially overwrite the pathological neural activity. The success of the DBS treatment is highly dependent upon the precision in the placement of the electrodes in the brain and the parameters of stimulation settings, including frequency, amplitude, and pulse width. The traditional DBS target for PD patients is subthalamic nucleus (STN), a gray matter (GM) structure. However, some white matter (WM) structures, such as hyperdirect pathway (HDP) and dentatorubrothalamic tract (DRT), have been proposed as alternatives.


The objective of the present study is to investigate the effect of WM stimulation for STN DBS treatment of PD patients by determining what proportion of WM activation is associated with therapeutic stimulation.


In this study, the motor symptom improvement of PD patients who underwent STN DBS treatment was clinically determined using the Movement Disorder Society-Unified Parkinson’s Disease Rating Scale (MDS-UPDRS). The internal capsule (IC) and STN voxels are known; however, the volume of tissue activation (VTA) voxels that are non-STN and non-IC are unknown. Therefore, a novel classification algorithm that can differentiate WM and GM voxels of patient specific volumes of tissue activation (VTAs) was developed. The relationship between the amount of WM within therapeutic VTAs and overall motor symptom improvement in PD patients was investigated.


The WM/GM classification algorithm was able to distinguish WM and GM voxels of patient specific therapeutic VTAs at an approximately 84.5% accuracy rate. Using the classification algorithm, the proportion of WM activation that is associated with therapeutic stimulation was calculated. Increase in WM activation had no significant effect on the overall clinical motor symptom improvement for PD patients.


The techniques used in this study are applicable to model other target sites which are composed of WM and GM. Hence, such a framework could be used to better characterize stimulation in the brain in regards to therapeutic and side effects associated with DBS.


Deep Brain Stimulation, Parkinson Disease, Neurological Movement Disorders, Neuromodulation, White Matter Stimulation, Machine Learning

Access Type

Honors Thesis (Bucknell Access Only)

Degree Type

Bachelor of Science in Biomedical Engineering


Biomedical Engineering

Minor, Emphasis, or Concentration


First Advisor

Karlo A. Malaga

Second Advisor

James Baish