Empowering Hand Rehabilitation with AI-Powered Gesture Recognition: A Study of an sEMG-Based System

Guo, Kai; Orban, Mostafa; Lu, Jingxin; Al-Quraishi, Maged S.; Yang, Hongbo; Elsamanty, Mahmoud

doi:10.3390/bioengineering10050557

Cited by 7 publications

(3 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, sEMG-based gesture recognition has attracted much attention due to its broad potential in the area of sign language, medical rehabilitation, virtual reality, and so on [ 25 ]. The approaches to tackle this classification problem can be categorized into conventional machine-learning-based approaches and deep-learning-based ones [ 5 , 7 , 9 , 11 , 26 ]. The former usually consists of three steps, including preprocessing sEMG signals, handcrafted feature extraction, and classification using the extracted features.…”

Section: Related Workmentioning

confidence: 99%

Improved Network and Training Scheme for Cross-Trial Surface Electromyography (sEMG)-Based Gesture Recognition

Dai,

Wong,

Kankanhali

et al. 2023

Bioengineering

View full text Add to dashboard Cite

To enhance the performance of surface electromyography (sEMG)-based gesture recognition, we propose a novel network-agnostic two-stage training scheme, called sEMGPoseMIM, that produces trial-invariant representations to be aligned with corresponding hand movements via cross-modal knowledge distillation. In the first stage, an sEMG encoder is trained via cross-trial mutual information maximization using the sEMG sequences sampled from the same time step but different trials in a contrastive learning manner. In the second stage, the learned sEMG encoder is fine-tuned with the supervision of gesture and hand movements in a knowledge-distillation manner. In addition, we propose a novel network called sEMGXCM as the sEMG encoder. Comprehensive experiments on seven sparse multichannel sEMG databases are conducted to demonstrate the effectiveness of the training scheme sEMGPoseMIM and the network sEMGXCM, which achieves an average improvement of +1.3% on the sparse multichannel sEMG databases compared to the existing methods. Furthermore, the comparison between training sEMGXCM and other existing networks from scratch shows that sEMGXCM outperforms the others by an average of +1.5%.

show abstract

Section: Related Workmentioning

confidence: 99%

Improved Network and Training Scheme for Cross-Trial Surface Electromyography (sEMG)-Based Gesture Recognition

Dai,

Wong,

Kankanhali

et al. 2023

Bioengineering

View full text Add to dashboard Cite

show abstract

“…Hand gesture recognition based on biosignals has recently become increasingly important for its potential to significantly improve the recovery and functionality of individuals with hand-related impairments [ 1 , 2 ]. Hand gestures are essential to human communication and interaction, and losing or impairing them can significantly impact the quality of life of individuals [ 1 , 2 , 3 ]. Surface electromyography (sEMG) signals have found widespread applications in various domains.…”

Section: Introductionmentioning

confidence: 99%

Comparative Study of sEMG Feature Evaluation Methods Based on the Hand Gesture Classification Performance

Hellara,

Barioul,

Sahnoun

et al. 2024

Sensors

View full text Add to dashboard Cite

Effective feature extraction and selection are crucial for the accurate classification and prediction of hand gestures based on electromyographic signals. In this paper, we systematically compare six filter and wrapper feature evaluation methods and investigate their respective impacts on the accuracy of gesture recognition. The investigation is based on several benchmark datasets and one real hand gesture dataset, including 15 hand force exercises collected from 14 healthy subjects using eight commercial sEMG sensors. A total of 37 time- and frequency-domain features were extracted from each sEMG channel. The benchmark dataset revealed that the minimum Redundancy Maximum Relevance (mRMR) feature evaluation method had the poorest performance, resulting in a decrease in classification accuracy. However, the RFE method demonstrated the potential to enhance classification accuracy across most of the datasets. It selected a feature subset comprising 65 features, which led to an accuracy of 97.14%. The Mutual Information (MI) method selected 200 features to reach an accuracy of 97.38%. The Feature Importance (FI) method reached a higher accuracy of 97.62% but selected 140 features. Further investigations have shown that selecting 65 and 75 features with the RFE methods led to an identical accuracy of 97.14%. A thorough examination of the selected features revealed the potential for three additional features from three specific sensors to enhance the classification accuracy to 97.38%. These results highlight the significance of employing an appropriate feature selection method to significantly reduce the number of necessary features while maintaining classification accuracy. They also underscore the necessity for further analysis and refinement to achieve optimal solutions.

show abstract

“…An investigation introduces a wearable system that integrates various assistive technologies for individuals with upper-limb impairments ( Guo et al, 2023 ; Orban et al, 2022 ). The system combines sensory components, haptic feedback mechanisms, orthotic devices, and robotics to facilitate forearm lifting and enhance grasping capabilities.…”

Section: Introductionmentioning

confidence: 99%

Soft pneumatic muscles for post-stroke lower limb ankle rehabilitation: leveraging the potential of soft robotics to optimize functional outcomes

Orban,

Guo,

Yang

et al. 2023

Front. Bioeng. Biotechnol.

Self Cite

View full text Add to dashboard Cite

Introduction: A soft pneumatic muscle was developed to replicate intricate ankle motions essential for rehabilitation, with a specific focus on rotational movement along the x-axis, crucial for walking. The design incorporated precise geometrical parameters and air pressure regulation to enable controlled expansion and motion.Methods: The muscle’s response was evaluated under pressure conditions ranging from 100-145 kPa. To optimize the muscle design, finite element simulation was employed to analyze its performance in terms of motion range, force generation, and energy efficiency. An experimental platform was created to assess the muscle’s deformation, utilizing advanced techniques such as high-resolution imaging and deep-learning position estimation models for accurate measurements. The fabrication process involved silicone-based materials and 3D-printed molds, enabling precise control and customization of muscle expansion and contraction.Results: The experimental results demonstrated that, under a pressure of 145 kPa, the y-axis deformation (y-def) reached 165 mm, while the x-axis and z-axis deformations were significantly smaller at 0.056 mm and 0.0376 mm, respectively, highlighting the predominant elongation in the y-axis resulting from pressure actuation. The soft muscle model featured a single chamber constructed from silicone rubber, and the visually illustrated and detailed geometrical parameters played a critical role in its functionality, allowing systematic manipulation to meet specific application requirements.Discussion: The simulation and experimental results provided compelling evidence of the soft muscle design’s adaptability, controllability, and effectiveness, thus establishing a solid foundation for further advancements in ankle rehabilitation and soft robotics. Incorporating this soft muscle into rehabilitation protocols holds significant promise for enhancing ankle mobility and overall ambulatory function, offering new opportunities to tailor rehabilitation interventions and improve motor function restoration.

show abstract

Empowering Hand Rehabilitation with AI-Powered Gesture Recognition: A Study of an sEMG-Based System

Cited by 7 publications

References 37 publications

Improved Network and Training Scheme for Cross-Trial Surface Electromyography (sEMG)-Based Gesture Recognition

Improved Network and Training Scheme for Cross-Trial Surface Electromyography (sEMG)-Based Gesture Recognition

Comparative Study of sEMG Feature Evaluation Methods Based on the Hand Gesture Classification Performance

Soft pneumatic muscles for post-stroke lower limb ankle rehabilitation: leveraging the potential of soft robotics to optimize functional outcomes

Contact Info

Product

Resources

About