Related Experiment Video
Updated: Aug 1, 2025
05:19
Using Neuron Spiking Activity to Trigger Closed-Loop Stimuli in Neurophysiological Experiments
Published on: November 12, 2019
7.1K
Landscape and future directions of machine learning applications in closed-loop brain stimulation
Anirudha S Chandrabhatla1, I Jonathan Pomeraniec2,3, Taylor M Horgan1
1School of Medicine, University of Virginia Health Sciences Center, Charlottesville, VA, 22903, USA.
NPJ Digital Medicine
|April 27, 2023
Summary
Machine learning (ML) enhances closed-loop brain stimulation (BStim) for neurological disorders. ML algorithms adapt stimulation in real-time, improving treatment for movement disorders, epilepsy, and neuropsychiatric conditions.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Computational Neuroscience
Background:
- Brain stimulation (BStim) uses implanted electrodes for neurological disorders, expanding beyond movement disorders to neuropsychiatric conditions.
- Traditional open-loop BStim delivers constant stimulation, while advanced closed-loop systems dynamically adjust based on neural biomarkers.
- Machine learning (ML) is crucial for developing closed-loop BStim systems by analyzing neural activity and adapting stimulation.
Purpose of the Study:
- To investigate the role of ML in developing closed-loop BStim systems for epilepsy, movement disorders, and neuropsychiatric disorders.
- To understand how ML algorithms are used to predict disease symptoms and modulate stimulation in real-time.
- To assess the performance of ML-driven closed-loop BStim systems compared to traditional methods.
Main Methods:
- Literature search of the US National Library of Medicine PubMed database.
- Analysis of studies employing both neural and non-neural network ML algorithms for closed-loop BStim.
- Categorization of research based on neurological disorder type (movement, epilepsy, neuropsychiatric).
Main Results:
- ML algorithms, both neural and non-neural, have been successfully applied to create closed-loop BStim systems.
- Closed-loop systems using ML demonstrate performance comparable to open-loop systems.
- Research for well-understood disorders focuses on ML model refinement for signal classification, while neuropsychiatric research targets biomarker identification for ML models.
Conclusions:
- ML is a vital component in advancing closed-loop brain stimulation for diverse neurological conditions.
- Closed-loop BStim systems powered by ML offer adaptive, patient-specific, and energy-efficient therapeutic modulation.
- Future research in neuropsychiatric disorders will focus on identifying robust neural biomarkers for ML-driven symptom detection and severity stratification.
