A Generalized Framework for the Study of Spinal Motor Neurons Controlling the Human Hand During Dynamic Movements

Cakici, Andre L.; Oßwald, Marius; Oliveira, Daniela Souza de; Braun, Dominik I.; Sîmpetru, Raul; Kinfe, Thomas M.; Eskofier, Bjoern M.; Vecchio, Alessandro Del

doi:10.1109/embc48229.2022.9870914

Cited by 6 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although the decomposition of the high-density EMG represents the most adequate solution since it mimics how the central nervous system encodes muscle forces [9], [48], [49], there are a large number of limitations in decoding a significant number of motor units during dynamic hand movement in real-time (see our previous conference paper [31] and tutorial article [26]) due to the high nonlinearities in the action potential shapes that are distorted by the contracting muscles [24]. Here we argue that machine learning, and more specifically deep learning, may represent the future for interfacing human movement with machines using the surface EMG signals.…”

Section: Discussionmentioning

confidence: 99%

“…• rock and roll sign We recorded three hand gestures (pointing, peace sign, and rock sign) as well as dynamic movements in order to investigate whether the model would have an easier time decoding the gestures or the dynamic movements and if there would be any discrepancies for the model to switching from individual finger actions to gestures. The acquisition system (detailed explanation in Cakici et al [31]) used 4 cameras distributed around a modular frame. The cameras recorded the movement of the hand simultaneously from different angles.…”

Section: • Peace Signmentioning

confidence: 99%

See 1 more Smart Citation

Proportional and Simultaneous Real-Time Control of the Full Human Hand From High-Density Electromyography

Sîmpetru¹,

März²,

Vecchio³

2023

Preprint

View full text Add to dashboard Cite

<p>Surface electromyography (sEMG) is a non-invasive technique that measures the electrical activity generated by the muscles using sensors placed on the skin. It has been widely used in the field of prosthetics and other assistive systems because of the physiological connection between muscle electrical activity and movement dynamics. However, most existing sEMG-based decoding algorithms show a limited number of detectable degrees of freedom that can be proportionally and simultaneously controlled in real-time, which limits the use of EMG in a wide range of applications, including prosthetics and other consumer-level applications (e.g., human/machine interfacing). In this work, we surpass the current state of the art by developing a new deep learning method that can decode and map the electrophysiological activity of the forearm muscles into proportional and simultaneous control of > 20 degrees of freedom of the human hand with real-time resolution and with latency within the neuromuscular delays (< 50 ms). We recorded the kinematics of the human hand during grasping, pinching, individual digit movements and three gestures at slow (0.5 Hz) and fast (0.75 Hz) movement speeds in healthy participants.</p> <p>We demonstrate that our neural network can predict the kinematics of the hand in real-time at a constant 32 predictions per second. To achieve this, we employed transfer learning and created a prediction smoothing algorithm for the output of the neural network that reconstructed the full geometry of the hand in three-dimensional Cartesian space in real-time. Our results demonstrate that high-density EMG signals from the forearm muscles contain almost all the information that is needed to predict the kinematics of the human hand. The proposed method has the capability of predicting the full kinematics of the human hand in an unprecedented way and with real-time resolution with immediate translational impact in subjects with motor impairments. </p>

show abstract

Section: Discussionmentioning

confidence: 99%

Section: • Peace Signmentioning

confidence: 99%

Proportional and Simultaneous Real-Time Control of the Full Human Hand From High-Density Electromyography

Sîmpetru¹,

März²,

Vecchio³

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The first dataset served as the control group and consists of 13 young, uninjured adult participants (age 25.9 ± 2.8 years) (data also used in [13]- [15]).…”

Section: Datasetsmentioning

confidence: 99%

Identification of Spared and Proportionally Controllable Hand Motor Dimensions in Motor Complete Spinal Cord Injuries Using Latent Manifold Analysis

Sîmpetru,

Souza de Oliveira,

Ponfick

et al. 2024

Preprint

View full text Add to dashboard Cite

The loss of bilateral hand function is a debilitating challenge for millions of individuals that suffered a motor-complete spinal cord injury (SCI). We have recently demonstrated in eight tetraplegic individuals the presence of highly functional spared spinal motor neurons in the extrinsic muscles of the hand that are still capable of generating proportional flexion and extension signals. In this work, we hypothesized that an artificial intelligence (AI) system could automatically learn the spared electromyographic (EMG) patterns that encode the attempted movements of the paralyzed digits. We constrained the AI to continuously output the attempted movements in the form of a digital hand so that this signal could be used to control any assistive system (e.g., exoskeletons, electrical stimulation). We trained a convolutional neural network using data from 13 uninjured (control) participants and 8 motor-complete tetraplegic participants to study the latent space learned by the AI. Our model can automatically differentiate between eight different hand movements, including individual finger flexions, grasps, and pinches, achieving a mean accuracy of 98.3% within the SCI group. Moreover, the model could distinguish with 100% accuracy whether a participant had an injury or not, and it could also facilitate proportional control of certain movements after the injury. Analysis of the latent space of the model revealed that proportionally controllable movements exhibited an elliptical path, while movements lacking proportional control followed a chaotic trajectory. We found that proportional control of a movement can only be correctly estimated if the latent space embedding of the movement follows an elliptical path (correlation = 0.73; p < 0.001). These findings emphasize the reliability of the proposed system for closed-loop applications that require an accurate estimate of the spinal cord motor output.

show abstract

“…Recent advancements in sEMG, particularly high-density sEMG (HD-sEMG), have allowed for accurate extraction of individual motor units using techniques such as convolutive kernel compensation (CKC) and fast independent component analysis (FastICA) (3)(4)(5)(6)(7)(8)(9). The characteristics of motor units have been investigated in both isometric and dynamic movements of the hand (4,8,(10)(11)(12)(13), with some studies showing the identification of unique motor units specific to certain movement patterns (14). Real-time decomposition of sEMG signals into motor unit firings, also known as online decomposition, has been successfully applied using convolutive blind source separation (BSS) techniques and gated recurrent units (GRU) (15)(16)(17)(18)(19).…”

Section: Introductionmentioning

confidence: 99%

NeurOne: High-performance Motor Unit-Computer Interface for the Paralyzed

Braun,

Souza de Oliveira,

Bayer

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

We have recently demonstrated that humans with motor-and-sensory complete cervical spinal cord injury (SCI) can modulate the activity of spared motor neurons that control the movements of paralyzed muscles. These motor neurons still receive highly functional cortical inputs that proportionally control flexion and extension movements of the paralyzed hand digits. In this study, we report a series of longitudinal experiments in which subjects with motor complete SCI received motor unit feedback from NeurOne. NeurOne is a software that realizes super-fast digitalization of motor neuron spiking activity (32 frames/s) and control of these neural ensembles through a physiological motor unit twitch model that enables intuitive brain-computer interactions closely matching the voluntary force modulation of healthy hand digits. We asked the subjects (n=3, 3-4 laboratory visits) to match a target displayed on a monitor through a cursor that was controlled by the modulation of the recruitment and rate coding of the spared motor units using a motor unit twitch model. The attempted movements of the paralyzed hands involved grasping and hand digit extension/flexion. The target cursor was scaled in a way that the subjects could increase or decrease feedback by either recruiting or derecruiting motor units, or by modulating the instantaneous discharge rate. The subjects learned to control the motor unit output with high levels of accuracy across different target intensities up to the maximal achievable discharge rate. Indeed, the high-performance motor output was surprisingly stable in a similar way as healthy subjects modulated the muscle force output recorded by a dynamometer. Therefore, NeurOne enables tetraplegic individuals an intuitive control of the paralyzed muscles through a digital neuromuscular system.Significance StatementOur study demonstrates the remarkable ability of individuals with complete cervical spinal cord injuries to modulate spared motor neurons and control paralyzed muscles. Utilizing NeurOne, a software, we enabled intuitive brain-computer interactions by digitalizing motor neuron spiking activity and employing a motor unit twitch model. Through this interface, tetraplegic individuals achieved high levels of accuracy and proportional control which closely resembled motor function in intact humans. NeurOne provides a promising digital neuromuscular interface, empowering individuals to control assistive devices super-fast and intuitive. This study signifies an important advancement in enhancing motor function and improving the quality of life for those with spinal cord injuries.

show abstract

A Generalized Framework for the Study of Spinal Motor Neurons Controlling the Human Hand During Dynamic Movements

Cited by 6 publications

References 14 publications

Proportional and Simultaneous Real-Time Control of the Full Human Hand From High-Density Electromyography

Proportional and Simultaneous Real-Time Control of the Full Human Hand From High-Density Electromyography

Identification of Spared and Proportionally Controllable Hand Motor Dimensions in Motor Complete Spinal Cord Injuries Using Latent Manifold Analysis

NeurOne: High-performance Motor Unit-Computer Interface for the Paralyzed

Contact Info

Product

Resources

About