Advanced Deep Brain Stimulation programming based on surface and intracranial electrophysiological readings

DBS is an efficacious procedure that allows for decreases in medication quantity, motor improvement and lessened adverse effects in patients with advanced Parkinson´s disease (Krack, Volkmann and Tinkhauser, 2019). With this treatment, an electric field is applied near a brain target, usually the subthalamic nucleus, using a variety of parameters such as voltage, pulse width or frequency. While clinical response is generally very good, some patients do not benefit completely from the procedure, while others experience surgery related side effects and there is currently no established method to predict the onset of complications, or select the patients who will show greater improvements.
In recent years, technological advancements in impulse generators have allowed not only to stimulate small brain regions, but also to record brain activity in real time, which can be used to adapt the stimulation to the patient’s needs (Bouthour et al., 2019). The use of combined surface recordings via electroencephalography (EEG) . Thus, harnessing the power of these two new technologies can shift the paradigm of the current blind, trial and error DBS programming to a future of adaptive, patient specific neuromodulation.

Important Notice: All clinical and neuroimaging data are highly sensitive and subject to strict confidentiality and intellectual property protocols. Any manipulation of the data outside the scope of the initial study protocol must be communicated to the project supervisor and will be individually evaluated.

Project Goals
The student will join a collaborative project between the UPF and the Movement Disorders Unit. It will collaborate with a translational group of researchers with expertise in both clinical and engineering fields. The specific goals of the project are:

  1. Management and analysis of intracranial local field potential (LFP) and surface EEG recordings in patients undergoing DBS. The student will learn how to acquire, preprocess, and visualize LFP and EEG data, and apply basic signal processing techniques to extract relevant neural features.
  2. Integration of clinical data with neurophysiological recordings. The student will learn to synchronize, curate, and interpret LFP and EEG signals in the context of clinical events, stimulation parameters, and patient specific factors.
  3. Identification of neural biomarkers for adaptive DBS. The student will explore how features derived from LFP and EEG can be used to guide DBS programming and potentially enable real time, closed loop neuromodulation.
  4. Assessment of individual variability. The project will involve comparing neural signatures and biomarker patterns across patients to better understand differences in DBS response and optimize personalized stimulation strategies.

Required Computational Skills

The student will have close support and supervision from the clinical technical team for data analysis and biomedical reasoning. Preferred experience or interest in:

  • Programming Languages: MATLAB (required), Python (desirable).
  • Image Processing: Experience with Lead-DBS, MATLAB.
  • Data Analysis: Use of statistical packages, supervised machine learning (classification)
  • Other Skills: Development or adaptation of signal processing algorithms, interest in computational neuroscience.

 

Supervisor: Ignacio Aracil, Laura Pérez Carasol