Search research articles

Search research articles

Related Experiment Video

Updated: Jun 8, 2025

Author Spotlight: Comparative Imaging of Neural Activity in Awake and Freely Moving States

Author Spotlight: Comparative Imaging of Neural Activity in Awake and Freely Moving States

Published on: January 19, 2024

Decoding multi-limb movements from two-photon calcium imaging of neuronal activity using deep learning.

Seungbin Park¹, Megan Lipton¹, Maria C Dadarlat¹

¹Weldon School of Biomedical Engineering, Purdue University, West Lafayette, IN 47906, United States of America.

Journal of Neural Engineering

|November 7, 2024

Summary

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

A shared multi-feature population code for sensory reliability across mouse visual cortex.

bioRxiv : the preprint server for biology·2026

Same author

Population-level encoding of somatosensation in mouse sensorimotor cortex.

Journal of neurophysiology·2026

Same author

Inhibitory network predicts microstimulation-induced circuit changes in the awake mammalian cortex.

bioRxiv : the preprint server for biology·2026

Same author

Flow stimuli reveal ecologically appropriate responses in mouse visual cortex.

Proceedings of the National Academy of Sciences of the United States of America·2018

Same author

Locomotion Enhances Neural Encoding of Visual Stimuli in Mouse V1.

The Journal of neuroscience : the official journal of the Society for Neuroscience·2017

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

Deep learning decodes multi-limb movements from slow optical brain recordings. This advance in brain-machine interfaces (BMIs) uses a novel recurrent neural network to interpret neural activity for controlling prosthetic limbs.

Area of Science:

Neuroscience
Biomedical Engineering
Machine Learning

Background:

Brain-machine interfaces (BMIs) aim to restore function after neural injury.
Decoding movement intention from neural activity is crucial for BMIs.
Two-photon (2p) calcium imaging offers high-resolution neural recordings but faces challenges with slow sampling rates for decoding fast movements.

Purpose of the Study:

To apply deep learning to decode multi-limb movements from 2p calcium imaging data.
To overcome the challenge of relating slow optical imaging data to fast behaviors.
To establish groundwork for controlling multiple limbs using optical BMIs.

Main Methods:

Developed a recurrent encoder-decoder network (LSTM-encdec) with an output longer than the input.

Keywords:

brain–machine interface deep learning multi-limb neural decoding optical brain–machine interface sensorimotor two-photon calcium imaging

More Related Videos

Two-photon Calcium Imaging in Neuronal Dendrites in Brain Slices

Two-photon Calcium Imaging in Neuronal Dendrites in Brain Slices

Published on: March 15, 2018

Multi-Fiber Photometry to Record Neural Activity in Freely-Moving Animals

Multi-Fiber Photometry to Record Neural Activity in Freely-Moving Animals

Published on: October 20, 2019

Related Experiment Videos

Last Updated: Jun 8, 2025

Author Spotlight: Comparative Imaging of Neural Activity in Awake and Freely Moving States

Author Spotlight: Comparative Imaging of Neural Activity in Awake and Freely Moving States

Published on: January 19, 2024

Two-photon Calcium Imaging in Neuronal Dendrites in Brain Slices

Two-photon Calcium Imaging in Neuronal Dendrites in Brain Slices

Published on: March 15, 2018

Multi-Fiber Photometry to Record Neural Activity in Freely-Moving Animals

Multi-Fiber Photometry to Record Neural Activity in Freely-Moving Animals

Published on: October 20, 2019

Applied the LSTM-encdec model to analyze 2p calcium imaging data from sensorimotor areas of running mice.

Utilized interpretability measures to validate decoding accuracy.

Main Results:

The LSTM-encdec model accurately decoded information about all four limbs (contralateral and ipsilateral front and hind limbs).
Decoding was achieved from calcium imaging data recorded in a single cortical hemisphere.
The approach demonstrated the feasibility of decoding complex, multi-limb movements.

Conclusions:

Deep learning, specifically the LSTM-encdec network, can effectively decode multi-limb movements from slow 2p calcium imaging data.
This method enhances the utility of optical imaging for brain-machine interfaces.
The findings pave the way for developing next-generation optical BMIs capable of controlling multiple limbs.