Short projects / EPFL lab immersions
Development of brain decoding algorithms to predict locomotor deficits in Parkinson’s patients
Background & Motivation
We are pursuing the development of novel therapies that employ closed-loop deep brain stimulation (DBS), in conjunction with spinal cord stimulation (SCS), to help alleviate deficits of gait and balance in patients with Parkinson’s disease (PD).
Our laboratory has extensive expertise in monitoring and modulating gait patterns in real-time based on online feedback of locomotor performance. Over the past 10 years, we established a unique, high-resolution platform combining real-time tracking of whole-body kinematics, recordings of muscle activity patterns for multiple synergistic muscles from both legs, along with cortical brain signals (EEG).
In addition, clinical advances in implantable devices for stimulation of Basal Ganglia circuits in Parkinson’s patients have opened the opportunity to study the function of these circuits during behavioral motor tasks. We have the unique possibility to record local field potentials (LFPs) from deep brain structures, which are known to become dysfunctional in PD and to exhibit pathological activity patterns that are at the origin of well-known motor symptoms.
Our aim is to establish a comprehensive description of gait deficits in PD in order to better understand the disease, to uncover the neural correlates underlying gait problems, and to develop decoding algorithms that can predict and trigger corrective stimulation patterns in closed-loop.
Goal of the project
Gait impairments in Parkinson’s disease are not restricted to leg movements, but also heavily affect upper- and lower-limb coordination, trunk posture and balance. The design of methodologies aimed to address such deficits hence require a comprehensive characterization of the phenotype of each patient, closely linked to what happens in the brain when symptoms arise. This, in turn, will allow us to predict them and help us address them.
This project will aim to develop models and decoding algorithms that will employ (i) basal ganglia LFP and (ii) cortical EEG to predict specific aspects of leg motor function, such as the timing of gait, propulsion effort during gait and, if possible, global gait deficits — The project will need to identify what leg variables are most relevant for each patient, develop brain decoders that help predict them, implement them within a real-time platform and test their performance during behavioral motor tasks with patients.
Tasks & Environment
– Participate in experiments with patients, help refine recording protocols and analytical methodologies.
– Analyze brain signals (EEG, LFP) and locomotor patterns (EMG, Kinematics) — Finding neural correlates of motor performance
– Develop decoding algorithms that help predict key aspects of gait (e.g., timing of gait events, measures of effort). Test the performance of such algorithms off-line
– Implement such decoders within a real-time recording and stimulation platform, and test during behavioral experiments with subjects.
The student is expected to be proficient in Matlab, have experience in Machine Learning, Signal processing and an honest scientific interest in Neuroscience — Previous experience with real-time software and/or in working in neural engineering-related projects would be a plus.
The project will be supervised by Dr Eduardo Martin Moraud and Yohann Thenaisie at the University hospital in Lausanne (CHUV). The student will have the possibility to closely interact with colleagues of the Department of Clinical Neuroscience at CHUV and the Centre of Neuroprosthetics at EPFL (Group of Prof. Courtine at Campus Biotech)
Duration: 4-6 months
Contact: [email protected]
Lozano et al. (2012), Probing and Regulating Dysfunctional Circuits Using Deep Brain Stimulation. Neuron
Tan et al. (2016), Decoding gripping force based on local field potentials recorded from subthalamic nucleus in humans. eLife
Fasano et al. (2015), Axial disability and deep brain stimulation in patients with Parkinson disease, Nature Reviews in Neurology
Collomb-Clerc et al. (2015), Effects of deep brain stimulation on balance and gait in patients with Parkinson’s disease: A systematic neurophysiological review. Clinical Neurophysiology.
Fischer et al. (2018), Alternating Modulation of Subthalamic Nucleus Beta Oscillations during Stepping. Journal of Neuroscience