Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Multicenter real world validation of on device single lead ECG parameters for remote cardiac assessment.

NPJ cardiovascular health·2026

Same author

PCSK9 promotes atherosclerosis progression through the FOXO3a autophagy signaling pathway.

Frontiers in cardiovascular medicine·2026

Same author

Artificial Intelligence-Enhanced Electrocardiogram for Longitudinal Evaluation of Right Ventricular Dysfunction After Acute Pulmonary Embolism: Rational and Design of CURES-CARE Study.

Pulmonary circulation·2026

Same author

Reconstructing 12-lead ECG from 3-lead ECG using variational autoencoder to improve cardiac disease detection of wearable ECG devices.

PLOS digital health·2026

Same author

ECGomics: An Open Platform for AI-ECG Digital Biomarker Discovery.

Health data science·2026

Same author

SpiroLLM: Finetuning pretrained LLMs to understand spirogram time series with clinical validation in COPD reporting.

PLOS digital health·2026

Search research articles

Related Experiment Video

Updated: May 3, 2026

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

Published on: March 10, 2011

Using reinforcement learning to provide stable brain-machine interface control despite neural input reorganization.

Eric A Pohlmeyer¹, Babak Mahmoudi², Shijia Geng¹

¹Department of Biomedical Engineering, University of Miami, Coral Gables, Florida, United States of America.

|February 6, 2014

Summary

This summary is machine-generated.

This study introduces an adaptive brain-machine interface (BMI) using reinforcement learning. The novel system allows for stable neural control of robotic devices, even with significant changes in brain activity.

More Related Videos

Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface

Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface

Published on: May 8, 2021

Brain-Computer Interface-controlled Upper Limb Robotic System for Enhancing Daily Activities in Stroke Patients

Brain-Computer Interface-controlled Upper Limb Robotic System for Enhancing Daily Activities in Stroke Patients

Published on: April 18, 2025

Related Experiment Videos

Last Updated: May 3, 2026

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

Published on: March 10, 2011

Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface

Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface

Published on: May 8, 2021

Brain-Computer Interface-controlled Upper Limb Robotic System for Enhancing Daily Activities in Stroke Patients

Brain-Computer Interface-controlled Upper Limb Robotic System for Enhancing Daily Activities in Stroke Patients

Published on: April 18, 2025

Area of Science:

Neuroscience
Robotics
Machine Learning

Background:

Brain-machine interface (BMI) systems enable direct neural control of external devices.
Current BMI systems face challenges in calibration, long-term stability, and adaptability to neural changes.
Static decoding algorithms struggle with non-stationary neural input.

Purpose of the Study:

To develop an adaptive BMI controller using reinforcement learning.
To enable stable and responsive neural decoding with minimal user calibration.
To address challenges posed by neural reorganizations in long-term BMI use.

Main Methods:

Implemented an actor-critic reinforcement learning architecture for BMI control.
Utilized binary evaluative feedback for learning robot arm control.
Tested the system with two marmoset monkeys in a reaching task.

Main Results:

The adaptive BMI (RLBMI) successfully controlled a robotic arm from brain states.
RLBMI demonstrated rapid learning with random initial conditions and minimal feedback.
The system maintained stable control over multiple weeks and adapted to neural input perturbations (halving/doubling neuron space).

Conclusions:

Reinforcement learning provides an effective adaptive controller for brain-machine interfaces.
The developed RLBMI overcomes limitations of static decoding algorithms.
This approach offers a promising path for robust and user-friendly BMI systems in daily living applications.