Deep brain stimulation (DBS) is a neurosurgical procedure that has been commonly used to treat a variety of conditions such as Parkinson’s Disease, Essential Tremor (ET), and epilepsy, among others. Briefly, DBS is a procedure where a surgeon implants electrodes into targeted areas of the brain. Despite being a well-established therapy, DBS has a high revision rate. The most common reason is improper electrode placement, which can be caused by placing the electrodes in a different location than intended or using a suboptimal stimulation site. The research areas associated with finding solutions to these problems are generally referred to as intraoperative localization and pre-operative targeting. In this thesis, Diffusion Tensor Imaging (DTI) data is used to investigate patient-specific approaches to ultimately reduce the revision rate of DBS.

The first study of this thesis explores a novel method of lead localization through both single electrode and dual electrode impedance measurements. First, it provides a theoretical basis for each impedance-based method using real patient data on theoretical trajectories. Second, the study investigates the relationship between DTI data and simulated impedance measurements on hypothetical DBS trajectories. Third, the study shows the potential value of monopolar simulated impedance measurements by evaluating its viability for localization. Lastly, this study shows that real monopolar impedance measured intraoperatively can match simulated impedance profiles. In full, we present the basis for a computationally efficient and patient specific method for intra operative lead localization through DTI-based impedance measurements.

The second study in this thesis retrospectively evaluates the potential of the motor hyperdirect pathway (HDP) as a target for DBS. The motor HDP connects the subthalamic nucleus (STN), a traditional DBS target, directly to the motor regions of the brain. This study uses DTI to create patient-specific models of the motor HDP to determine if it is a viable alternative target for DBS. However, we find that while the study explored several potentially useful applications of DTI for targeting, higher activation of motor HDP fibers was not associated with better patient outcomes.

This thesis presents two methods for improved electrode placement during DBS. The first study provides a foundation for an impedance-based localization scheme that has the potential to be applied not only during STN-targeted DBS (STN-DBS) but across a broad range of other stereotactic neurosurgeries. The second study found that increased motor HDP stimulation did not lead to a statistically significant increase in patient outcomes. However, this study presented a DTI-driven approach to targeting that can be used in future work. It can be used to understand the specific DT properties of targets that have found a statistically significant correlation to patient outcomes but were only conducted using MRI data. This is important as while MRI-based studies can find usable targets, they less explicitly elucidate the mechanism behind why the target was suitable.

DATE: Thursday, April 21, 2022
TIME: 2:00 PM
Zoom: https://umich.zoom.us/j/97612255859 (passcode: 840422)
Chair: Assoc. Prof. Parag Patil