Fusion of EEG and EMG signals for detecting pre-movement intention of sitting and standing in healthy individuals and patients with spinal cord injury DOI Creative Commons
Chenyang Li, Yuchen Xu, Tao Feng

и другие.

Frontiers in Neuroscience, Год журнала: 2025, Номер 19

Опубликована: Янв. 24, 2025

Introduction Rehabilitation devices assist individuals with movement disorders by supporting daily activities and facilitating effective rehabilitation training. Accurate early motor intention detection is vital for real-time device applications. However, traditional methods of often rely on single-mode signals, such as EEG or EMG alone, which can be limited low signal quality reduced stability. This study proposes a multimodal fusion method based EEG–EMG functional connectivity to detect sitting standing intentions before execution, enabling timely intervention reducing latency in devices. Methods Eight healthy subjects five spinal cord injury (SCI) patients performed cue-based sit-to-stand stand-to-sit transition tasks while data were recorded simultaneously. We constructed networks using epochs from the 1.5-s period prior onset. Pairwise spatial filters then designed extract discriminative network topologies. Each filter paired support vector machine classifier classify future movements into one three classes: sit-to-stand, stand-to-sit, rest. The final prediction was determined majority voting scheme. Results Among investigated—coherence, Pearson correlation coefficient mutual information (MI)—the MI-based showed highest decoding performance (94.33%), outperforming both (73.89%) (89.16%). robustness further validated through fatigue training experiment subjects. achieved 92.87% accuracy during post-fatigue stage, no significant difference compared pre-fatigue stage ( p > 0.05). Additionally, proposed pre-movement windows comparable trans-movement 0.05 pre- stages). For SCI patients, improved accuracy, achieving 87.54% single- modality (EEG: 83.03%, EMG: 84.13%), suggesting that could promising practical Conclusion Our results demonstrated significantly enhances detecting human intentions. By intentions, this holds potential offer more accurate interventions within systems.

Язык: Английский

Fusion of EEG and EMG signals for detecting pre-movement intention of sitting and standing in healthy individuals and patients with spinal cord injury DOI Creative Commons
Chenyang Li, Yuchen Xu, Tao Feng

и другие.

Frontiers in Neuroscience, Год журнала: 2025, Номер 19

Опубликована: Янв. 24, 2025

Introduction Rehabilitation devices assist individuals with movement disorders by supporting daily activities and facilitating effective rehabilitation training. Accurate early motor intention detection is vital for real-time device applications. However, traditional methods of often rely on single-mode signals, such as EEG or EMG alone, which can be limited low signal quality reduced stability. This study proposes a multimodal fusion method based EEG–EMG functional connectivity to detect sitting standing intentions before execution, enabling timely intervention reducing latency in devices. Methods Eight healthy subjects five spinal cord injury (SCI) patients performed cue-based sit-to-stand stand-to-sit transition tasks while data were recorded simultaneously. We constructed networks using epochs from the 1.5-s period prior onset. Pairwise spatial filters then designed extract discriminative network topologies. Each filter paired support vector machine classifier classify future movements into one three classes: sit-to-stand, stand-to-sit, rest. The final prediction was determined majority voting scheme. Results Among investigated—coherence, Pearson correlation coefficient mutual information (MI)—the MI-based showed highest decoding performance (94.33%), outperforming both (73.89%) (89.16%). robustness further validated through fatigue training experiment subjects. achieved 92.87% accuracy during post-fatigue stage, no significant difference compared pre-fatigue stage ( p > 0.05). Additionally, proposed pre-movement windows comparable trans-movement 0.05 pre- stages). For SCI patients, improved accuracy, achieving 87.54% single- modality (EEG: 83.03%, EMG: 84.13%), suggesting that could promising practical Conclusion Our results demonstrated significantly enhances detecting human intentions. By intentions, this holds potential offer more accurate interventions within systems.

Язык: Английский

Процитировано

0