NSF DARE—transforming modeling in neurorehabilitation: a patient-in-the-loop framework

Journal of NeuroEngineering and Rehabilitation, Journal Year: 2024, Volume and Issue: 21(1)

Published: Feb. 13, 2024

Abstract In 2023, the National Science Foundation (NSF) and Institute of Health (NIH) brought together engineers, scientists, clinicians by sponsoring a conference on computational modelling in neurorehabiilitation. To facilitate multidisciplinary collaborations improve patient care, this perspective piece we identify where how can support neurorehabilitation. address where, developed patient-in-the-loop framework that uses multiple and/or continual measurements to update diagnostic treatment model parameters, type, prescription, with goal maximizing clinically-relevant functional outcomes. This has several key features: (i) it includes models, (ii) is clinically-grounded International Classification Functioning, Disability (ICF) involvement, (iii) or data over time, (iv) applicable range neurological neurodevelopmental conditions. how, state-of-the-art highlight promising avenues future research across realms sensorimotor adaptation, neuroplasticity, musculoskeletal, sensory & pain modelling. We also discuss both importance perform validation, as well challenges overcome when implementing models within clinical setting. The approach offers unifying guide collaboration between stakeholders field

Language: Английский

---

An EEG channel selection method for motor imagery based on Fisher score and local optimization

Journal of Neural Engineering, Journal Year: 2024, Volume and Issue: 21(3), P. 036030 - 036030

Published: June 1, 2024

Abstract Objective . Multi-channel electroencephalogram (EEG) technology in brain–computer interface (BCI) research offers the advantage of enhanced spatial resolution and system performance. However, this also implies that more time is needed data processing stage, which not conducive to rapid response BCI. Hence, it a necessary challenging task reduce number EEG channels while maintaining decoding effectiveness. Approach In paper, we propose local optimization method based on Fisher score for within-subject channel selection. Initially, extract common pattern characteristics signals different bands, calculate scores each these characteristics, rank them accordingly. Subsequently, employ finalize Main results On BCI Competition IV Dataset IIa, our selects an average 11 across four achieving accuracy 79.37%. This represents 6.52% improvement compared using full set 22 channels. self-collected dataset, similarly achieves significant 24.20% with less than half channels, resulting 76.95%. Significance explores importance combinations selection tasks reveals appropriately combining can further enhance quality The indicate model selected small higher two-class motor imagery classification tasks. Additionally, improves portability systems through combinations, offering potential development portable systems.

Language: Английский

---

Neural Plasticity in Sensorimotor Brain–Machine Interfaces

Annual Review of Biomedical Engineering, Journal Year: 2023, Volume and Issue: 25(1), P. 51 - 76

Published: Feb. 28, 2023

Brain–machine interfaces (BMIs) aim to treat sensorimotor neurological disorders by creating artificial motor and/or sensory pathways. Introducing pathways creates new relationships between input and output, which the brain must learn gain dexterous control. This review highlights role of learning in BMIs restore movement sensation, discusses how BMI design may influence neural plasticity performance. The close integration function influences both will be an essential consideration for bidirectional devices that function.

Language: Английский

---

Flexible intentions: An Active Inference theory

Frontiers in Computational Neuroscience, Journal Year: 2023, Volume and Issue: 17

Published: March 20, 2023

We present a normative computational theory of how the brain may support visually-guided goal-directed actions in dynamically changing environments. It extends Active Inference cortical processing according to which maintains beliefs over environmental state, and motor control signals try fulfill corresponding sensory predictions. propose that neural circuitry Posterior Parietal Cortex (PPC) compute flexible intentions-or plans from belief targets-to generate actions, we develop formalization this process. A proof-of-concept agent embodying visual proprioceptive sensors an actuated upper limb was tested on target-reaching tasks. The behaved correctly under various conditions, including static dynamic targets, different feedbacks, precisions, intention gains, movement policies; limit conditions were individuated, too. driven by intentions can thus behavior constantly environments, PPC might putatively host its core mechanism. More broadly, study provides basis for research end-to-end settings further advances mechanistic theories active biological systems.

Language: Английский

---

The mind–machine connection: adaptive information processing and new technologies promoting mental health in older adults

The Neuroscientist, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 19, 2025

The human brain demonstrates an exceptional adaptability, which encompasses the ability to regulate emotions, exhibit cognitive flexibility, and generate behavioral responses, all supported by neuroplasticity. Brain–computer interfaces (BCIs) employ adaptive algorithms machine learning techniques adapt variations in user’s activity, allowing for customized interactions with external devices. Older adults may experience decline, could affect learn new technologies such as BCIs, but both (human BCI) demonstrate adaptability their responses. is skilled at quickly switching between tasks regulating while BCIs can modify signal-processing accommodate changes activity. Furthermore, BCI participate knowledge acquisition; first one strengthens abilities through exposure experiences, second improves performance ongoing adjustment improvement. Current research seeks incorporate emotional states into systems improve user experience, despite regulation of brain. implementation older be more effective, inclusive, beneficial improving quality life. This review aims understanding brain–machine implications mental health adults.

Language: Английский

---

Deep Learning-Based Markerless Hand Tracking for Freely Moving Non-Human Primates in Brain–Machine Interface Applications

Electronics, Journal Year: 2025, Volume and Issue: 14(5), P. 920 - 920

Published: Feb. 26, 2025

The motor cortex of non-human primates plays a key role in brain–machine interface (BMI) research. In addition to recording cortical neural signals, accurately and efficiently capturing the hand movements experimental animals under unconstrained conditions remains challenge. Addressing this challenge can deepen our understanding application BMI behavior from both theoretical practical perspectives. To address issue, we developed deep learning framework that combines Yolov5 RexNet-ECA reliably detect joint positions freely moving at different distances using single camera. model simplifies setup procedure while maintaining high accuracy, with an average keypoint detection error less than three pixels. Our method eliminates need for physical markers, ensuring non-invasive data collection preserving natural subjects. proposed system exhibits accuracy ease use compared existing methods. By quickly acquiring spatiotemporal behavioral metrics, provides valuable insights into dynamic interplay between functions, further advancing

Language: Английский

---

MAIS: an in-vitro sandbox enables adaptive neuromodulation via scalable neural interfaces

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: March 17, 2025

Abstract Brain-machine interfaces (BMIs) predominantly rely on static digital architectures to decode biological neuronal networks, a paradigm that is incompatible with natural neural coding in the human brain 1–4 . Bridging this gap critical step combating dysfunction, enhancing functionality, and refining precision of neuroprosthetics 5 The integration organoids microelectrode array (MEA), as class BMIs, offers humanized vitro platform unique compatibility advantages for dynamic decoding. This study resolves biological-electronic encoding incompatibility organoid-MEA Integration through three progressive breakthroughs. First, human-machine hybrid agent developed newly proposed bioengineered couples together high-density MEAs computational chips, enabling closed-loop perturbation networks via exogenous signals. Second, plasticity-driven real-time tracking activity, we establish dynamically reconfigurable stimulation nodes self-align electrophysiological states organoids. exogenous-endogenous mismatch by implementing adaptation principles ensure spatially adaptive coordination. Finally, shared plasticity rules rather than centralized control, construct first scalable multi-agent interaction system (MAIS) demonstrate its real-world applications. Through designed scenarios pathological/normal network interaction, validate MAIS achieves stable cross-network embodies self-evolving sandbox which decoding bridges gaps between systems, providing foundational infrastructure human-centered interfaces.

Language: Английский

---

Neurocognitive and motor-control challenges for the realization of bionic augmentation

Nature Biomedical Engineering, Journal Year: 2022, Volume and Issue: 7(4), P. 344 - 348

Published: Sept. 1, 2022

Language: Английский

---

Altered resting-state brain function in endurance athletes

Cerebral Cortex, Journal Year: 2024, Volume and Issue: 34(3)

Published: Feb. 19, 2024

Previous research has confirmed significant differences in regional brain activity and functional connectivity between endurance athletes non-athletes. However, no studies have investigated the topological efficiency of network Here, we compared activities, connectivity, properties to explore basis associated with training. The results showed correlations Regional Homogeneity motor cortex, visual cerebellum, training intensity parameters. Alterations among inferior frontal gyrus cingulate were significantly correlated In addition, graph theoretical analysis revealed a reduction global athletes. This decline is mainly caused by decreased nodal local cerebellar regions. Notably, sensorimotor regions, such as precentral supplementary areas, still exhibit increased efficiency. study not only confirms improvement regions related training, but also offers novel insights into mechanisms through which undergo changes network.

Language: Английский

---

An Investigation of Manifold-Based Direct Control for a Brain-to-Body Neural Bypass

IEEE Open Journal of Engineering in Medicine and Biology, Journal Year: 2024, Volume and Issue: 5, P. 271 - 280

Published: Jan. 1, 2024

Objective: Brain-body interfaces (BBIs) have emerged as a very promising solution for restoring voluntary hand control in people with upper-limb paralysis. The BBI module decoding motor commands from brain signals should provide the user intuitive, accurate, and stable control. Here, we present preliminary investigation monkey of strategy based on direct coupling between activity intrinsic neural ensembles output variables, aiming at achieving ease learning long-term robustness. xmlns:xlink="http://www.w3.org/1999/xlink">Results: We identified an low-dimensional space (called manifold) capturing co-variation patterns monkey's associated to reach-to-grasp movements. then tested animal's ability directly computer cursor using cortical activation along manifold axes. By daily recalibrating only scaling factors, achieved rapid high performance simple, incremental 2D tasks over more than 12 weeks experiments. Finally, showed that this can be effectively coupled peripheral nerve stimulation trigger xmlns:xlink="http://www.w3.org/1999/xlink">Conclusions: These results represent proof concept manifold-based applications.

Language: Английский

---