What does this research mean for the field?

A fatigue-robust brain-muscle fusion network (FRFNet) utilizing degradation-aware adaptive fusion significantly improves EEG-EMG movement decoding accuracy under progressive EMG degradation compared to existing fusion methods. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To develop and validate a multimodal neural network (FRFNet) that preserves accurate hand movement intention recognition despite muscle fatigue-induced electromyography signal degradation.

September 14, 2026Frontiers in Human NeuroscienceOpen Access

FRFNet: a fatigue-robust EEG–EMG fusion network for hand movement intention recognition

View Full Paper

Ask AI

Bookmark

Authors

YCYingYu Cao SBShuoyu Bai ZZZhenxi Zhao

Discussion

Member takes

Overview

Computational study demonstrates fatigue-robust hand movement intention recognition across progressive muscle degradation levels, suggesting improved reliability for assistive human–machine...

Key Points

To develop and validate a multimodal neural network (FRFNet) that preserves accurate hand movement intention recognition despite muscle fatigue-induced electromyography signal degradation.
Engineered FRFNet comprising a multiscale temporal convolutional network for EEG, a dual-path fatigue disentanglement encoder for EMG, and dynamic weighted adaptive gating fusion.
Evaluated the framework on a public upper-limb EEG–EMG dataset subjected to synthetic fatigue degradation levels ranging from 10% to 90% across within-subject and leave-one-subject-out protocols.
FRFNet attained 60.0% accuracy at 90% EMG degradation in within-subject testing, exceeding E2FNet by 12.7, DCA Fusion by 15.2, and DMEFNet-adapted by 11.7 percentage points.
Under leave-one-subject-out evaluation at 90% degradation, the model achieved 57.0% accuracy, outperforming all five baseline methods.
The deployment architecture utilized 1.166 million parameters and demonstrated a single-trial forward latency of 5.321 ± 0.934 ms.

Cite This Study

Cao et al. (2026) studied this question.

synapsesocial.com/papers/6aa7b2c70926e14a848b14e9 https://doi.org/10.3389/fnhum.2026.1885665

View Full Paper

Ask AI

Bookmark

September 14, 2026Frontiers in Human NeuroscienceOpen Access

FRFNet: a fatigue-robust EEG–EMG fusion network for hand movement intention recognition

View Full Paper

Ask AI

Bookmark

Authors

YCYingYu Cao SBShuoyu Bai ZZZhenxi Zhao

Discussion

Member takes

Overview

Computational study demonstrates fatigue-robust hand movement intention recognition across progressive muscle degradation levels, suggesting improved reliability for assistive human–machine...

Key Points

To develop and validate a multimodal neural network (FRFNet) that preserves accurate hand movement intention recognition despite muscle fatigue-induced electromyography signal degradation.
Engineered FRFNet comprising a multiscale temporal convolutional network for EEG, a dual-path fatigue disentanglement encoder for EMG, and dynamic weighted adaptive gating fusion.
Evaluated the framework on a public upper-limb EEG–EMG dataset subjected to synthetic fatigue degradation levels ranging from 10% to 90% across within-subject and leave-one-subject-out protocols.
FRFNet attained 60.0% accuracy at 90% EMG degradation in within-subject testing, exceeding E2FNet by 12.7, DCA Fusion by 15.2, and DMEFNet-adapted by 11.7 percentage points.
Under leave-one-subject-out evaluation at 90% degradation, the model achieved 57.0% accuracy, outperforming all five baseline methods.
The deployment architecture utilized 1.166 million parameters and demonstrated a single-trial forward latency of 5.321 ± 0.934 ms.

Cite This Study

Cao et al. (2026) studied this question.

synapsesocial.com/papers/6aa7b2c70926e14a848b14e9 https://doi.org/10.3389/fnhum.2026.1885665

View Full Paper

Ask AI

Bookmark