Related Experiment Video
Updated: May 11, 2026
14:14
Targeting Neuronal Fiber Tracts for Deep Brain Stimulation Therapy Using Interactive, Patient-Specific Models
Published on: August 12, 2018
Pathological tremor prediction using surface electromyogram and acceleration: potential use in 'ON-OFF' demand driven
Ishita Basu1, Daniel Graupe, Daniela Tuninetti
1Department of Electrical and Computer Engineering, University of Illinois at Chicago (UIC), IL, USA. ibjuslc@gmail.com
Journal of Neural Engineering
|May 10, 2013
Summary
This study introduces a new method to predict pathological tremor in Parkinson's disease (PD) and essential tremor (ET) patients using surface electromyogram (sEMG) and acceleration data. The developed algorithm accurately forecasts tremor onset, paving the way for advanced closed-loop deep brain stimulation (DBS).
Area of Science:
- Biomedical Engineering
- Neuroscience
- Signal Processing
Background:
- Pathological tremor significantly impacts patients with Parkinson's disease (PD) and essential tremor (ET).
- Current treatments like deep brain stimulation (DBS) can be improved with predictive capabilities.
- Non-invasive monitoring methods offer potential for enhanced tremor management.
Purpose of the Study:
- To develop and validate a novel, non-invasive method for predicting the onset of pathological tremor.
- To utilize surface electromyogram (sEMG) and acceleration data for tremor prediction in PD and ET patients.
- To lay the groundwork for predictive closed-loop deep brain stimulation (DBS) systems.
Main Methods:
- Extracted spectral (Fourier, wavelet) and nonlinear time series (entropy, recurrence rate) parameters from sEMG and acceleration signals.
- Developed a tremor prediction algorithm based on these extracted signal parameters.
- Validated the algorithm on data from Parkinson's disease and essential tremor patients.
Main Results:
- The algorithm achieved 100% sensitivity in predicting tremor onset for all recorded trials in both PD and ET patients.
- Overall prediction accuracy reached 85.7% for ET trials and 80.2% for PD trials.
- Statistical analysis confirmed that prediction results significantly differed from random chance.
Conclusions:
- The developed tremor prediction algorithm shows high efficacy for both PD and ET.
- This predictive capability is crucial for designing next-generation non-invasive closed-loop DBS controllers.
- Optimized prediction allows for maximizing tremor-free intervals, reducing brain stimulation and battery consumption.
