Advancing Brain-Computer Interface System Performance in Hand Trajectory Estimation with NeuroKinect DOI Creative Commons
Sidharth Pancholi, Amita Giri

Published: July 30, 2023

<p>Brain-computer interface (BCI) technology enables direct communication between the brain and external devices, allowing individuals to control their environment using signals. However, existing BCI approaches face three critical challenges that hinder practicality effectiveness: a) time-consuming preprocessing algorithms, b) inappropriate loss function utilization, c) less intuitive hyperparameter settings. To address these limitations, we present NeuroKinect, an innovative deep-learning model for accurate reconstruction of hand kinematics electroencephalography (EEG) NeuroKinect is trained on Grasp Lift (GAL) tasks data with minimal pipelines, subsequently improving computational efficiency. A notable improvement introduced by utilization a novel function, denoted as LStat. This addresses discrepancy correlation mean square error in prediction. Furthermore, our study emphasizes scientific intuition behind parameter selection enhance accuracy. We analyze spatial temporal dynamics motor movement task employing event-related potential source localization (BSL) results. approach provides valuable insights into optimal selection, overall performance accuracy model. Our demonstrates strong correlations predicted actual movements, Pearson coefficients 0.92 (±0.015), 0.93 (±0.019), 0.83 (±0.018) X, Y, Z dimensions. The precision evidenced low squared errors (MSE) 0.016 (±0.001), 0.015 (±0.002), 0.017 (±0.005) dimensions, respectively. Overall, results demonstrate unprecedented real-time translation capability, making significant advancement field predicting from </p>

Language: Английский

Flexible brain–computer interfaces DOI
Xin Tang, Hao Shen, Siyuan Zhao

et al.

Nature Electronics, Journal Year: 2023, Volume and Issue: 6(2), P. 109 - 118

Published: Feb. 2, 2023

Language: Английский

Citations

188
Against cortical reorganisation DOI Creative Commons
Tamar R. Makin, John W. Krakauer

eLife, Journal Year: 2023, Volume and Issue: 12

Published: Nov. 21, 2023

Neurological insults, such as congenital blindness, deafness, amputation, and stroke, often result in surprising impressive behavioural changes. Cortical reorganisation, which refers to preserved brain tissue taking on a new functional role, is invoked account for these Here, we revisit many of the classical animal patient cortical remapping studies that spawned this notion reorganisation. We highlight empirical, methodological, conceptual problems call into doubt. argue appeal idea reorganisation attributable part way maps are empirically derived. Specifically, defined based oversimplified assumptions 'winner-takes-all', turn leads an erroneous interpretation what it means when appear change. Conceptually, interpreted circuit receiving novel input processing unrelated its original function. This implies neurons either pluripotent enough change they tuned or can computes. Instead more likely occur due potentiation pre-existing architecture already has requisite representational computational capacity pre-injury. be facilitated via Hebbian homeostatic plasticity mechanisms. Crucially, our revised framework proposes opportunities constrained throughout lifespan by underlying structural 'blueprint'. At no period, including early development, does cortex offer pluripotency. conclude distinct form plasticity, ubiquitously evoked with words 'take-over'' 'rewiring', not exist.

Language: Английский

Citations

62
Toward a Brain–Neuromorphics Interface DOI

Changjin Wan,

Mengjiao Pei,

Kailu Shi

et al.

Advanced Materials, Journal Year: 2024, Volume and Issue: 36(37)

Published: Feb. 10, 2024

Abstract Brain–computer interfaces (BCIs) that enable human–machine interaction have immense potential in restoring or augmenting human capabilities. Traditional BCIs are realized based on complementary metal‐oxide‐semiconductor (CMOS) technologies with complex, bulky, and low biocompatible circuits, suffer the energy efficiency of von Neumann architecture. The brain–neuromorphics interface (BNI) would offer a promising solution to advance BCI shape interactions machineries. Neuromorphic devices systems able provide substantial computation power extremely high energy‐efficiency by implementing in‐materia computing such as situ vector–matrix multiplication (VMM) physical reservoir computing. Recent progresses integrating neuromorphic components sensing and/or actuating modules, give birth afferent nerve, efferent sensorimotor loop, so on, which has advanced for future neurorobotics achieving sophisticated capabilities biological system. With development compact artificial spiking neuron bioelectronic interfaces, seamless communication between BNI bioentity is reasonably expectable. In this review, upcoming BNIs profiled introducing brief history neuromorphics, reviewing recent related areas, discussing advances challenges lie ahead.

Language: Английский

Citations

20
Bionic perception and transmission neural device based on a self-powered concept DOI Creative Commons
Kaixian Ba,

Guijiang Liu,

Guoliang Ma

et al.

Cell Reports Physical Science, Journal Year: 2024, Volume and Issue: 5(7), P. 102048 - 102048

Published: June 13, 2024

Differing from the traditional view that power frequency electric field is always considered a negative phenomenon, here we report concept uses directly as an energy source to generate sensing signals. To demonstrate this, integrated bionic perception and transmission nerve device (BPTND) based on developed. The BPTND can effectively simulate sensory nervous systems by integrating perception, recognition, functions detect transmit positional information of mechanical stimulation. Results shows advantages simple preparation, low cost, fast response, strong shape self-adaptation, mechanism, anti-interference ability. successfully applied automobile unmanned aerial vehicle control hydraulic quadruped robot leg motion control. This expected guide development various new forms modules in future.

Language: Английский

Citations

18
Memristor‐Based Bionic Tactile Devices: Opening the Door for Next‐Generation Artificial Intelligence DOI

Chuan Yang,

Hongyan Wang, Zelin Cao

et al.

Small, Journal Year: 2023, Volume and Issue: 20(19)

Published: Dec. 27, 2023

Bioinspired tactile devices can effectively mimic and reproduce the functions of human system, presenting significant potential in field next-generation wearable electronics. In particular, memristor-based bionic have attracted considerable attention due to their exceptional characteristics high flexibility, low power consumption, adaptability. These provide advanced wearability high-precision sensing capabilities, thus emerging as an important research area within bioinspired This paper delves into integration memristors with other controlling systems offers a comprehensive analysis recent advancements devices. incorporate artificial nociceptors flexible electronic skin (e-skin) category bio-inspired sensors equipped capabilities for sensing, processing, responding stimuli, which are expected catalyze revolutionary changes human-computer interaction. Finally, this review discusses challenges faced by terms material selection, structural design, sensor signal processing development intelligence. Additionally, it also outlines future directions application prospects these devices, while proposing feasible solutions address identified challenges.

Language: Английский

Citations

22
High-density electromyography for effective gesture-based control of physically assistive mobile manipulators DOI Creative Commons
Jehan Yang,

Kent Shibata,

Douglas J. Weber

et al.

npj Robotics, Journal Year: 2025, Volume and Issue: 3(1)

Published: Jan. 27, 2025

Language: Английский

Citations

0
Human culture is uniquely open-ended rather than uniquely cumulative DOI
Thomas J. H. Morgan, Marcus W. Feldman

Nature Human Behaviour, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 7, 2024

Language: Английский

Citations

3
Material Selection and Device Design of Scalable Flexible Brain‐Computer Interfaces: A Balance Between Electrical and Mechanical Performance DOI

Xinyi Lin,

X. Zhang,

Juntao Chen

et al.

Advanced Materials, Journal Year: 2025, Volume and Issue: unknown

Published: April 28, 2025

Abstract Brain‐computer interfaces (BCIs) hold the potential to revolutionize brain function restoration, enhance human capability, and advance our understanding of cognitive mechanisms by directly linking neural signals with hardware. However, mechanical mismatch between conventional rigid BCIs soft tissue limits long‐term interface stability. Next‐generation must achieve biocompatibility while maintaining high performance, enabling integration millions sensors within tissue‐level flexible soft, stable interfaces. Lithographic fabrication techniques provide scalable thin‐film electronics, but traditional electronic materials often fail meet unique requirements BCIs. This review examines selection device design for BCIs, starting an analysis intrinsic material properties—Young's modulus, electrical conductivity dielectric constant. It then explores electrode optimize circuits assess key factors. Next, correlation performance is analyzed guide design. Finally, recent advances in probes are reviewed, highlighting improvements signal quality, recording stability, scalability. focuses on scalable, lithography‐based aiming identify optimal designs long‐term, reliable recordings.

Language: Английский

Citations

0
Exploring Skill Generalization with an Extra Robotic Arm for Motor Augmentation DOI Creative Commons
Daniel J. L. L. Pinheiro, Giulia Dominijanni,

Francesca Paola Maenza

et al.

Advanced Intelligent Systems, Journal Year: 2025, Volume and Issue: unknown

Published: May 6, 2025

Recent research demonstrates that naïve users can be trained to perform complex motor tasks, including trimanual activities, using an extra robotic arm (XRA). While previous studies show task‐specific improvements with XRAs, it remains uncertain whether skills acquired in one task generalize others differing cognitive and demands. This study investigates multitasking training enhances performance on untrained tasks involving XRA. The combined biological functions (button pressing, slider movement, speech) XRA control via voluntary diaphragmatic modulation. Untrained include block manipulation concurrent keyboard typing. We compared the between a group only those completes beforehand. Training significantly improves ( t = 3.45, p 0.001). Additionally, users’ this is higher than when they used their limbs, demonstrating true functional augmentation 2.70, 0.021). However, no differences are observed groups typing 163.50, 0.880). These findings highlight need explore adaptive protocols enhancing XRA‐biological limb coordination for improved skill transfer across diverse environments.

Language: Английский

Citations

0
Impact of supplementary sensory feedback on the control and embodiment in human movement augmentation DOI Creative Commons
Mattia Pinardi, Matthew R. Longo, Domenico Formica

et al.

Communications Engineering, Journal Year: 2023, Volume and Issue: 2(1)

Published: Sept. 11, 2023

Abstract In human movement augmentation, the number of controlled degrees freedom could be enhanced by simultaneous and independent use supernumerary robotic limbs (SRL) natural ones. However, this poses several challenges, that mitigated encoding relaying SRL status. Here, we review impact supplementary sensory feedback on control embodiment SRLs. We classify main features analyse how they improve performance. report feasibility pushing body representation beyond morphology suggest gradual make multisensory incongruencies less disruptive. also highlight shared computational bases between motor contextualizing them within same theoretical framework. Finally, argue a shift towards long term experimental paradigms is necessary for successfully integrating embodiment.

Language: Английский

Citations

7