Article
Computer Science, Information Systems
Jia Zeng, Yu Zhou, Yicheng Yang, Jipeng Yan, Honghai Liu
Summary: This study compared the performance of sEMG and AUS in hand gesture recognition tasks under muscle fatigue conditions, and found that AUS signal has better fatigue robustness than sEMG signal.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2022)
Article
Engineering, Biomedical
Xiaoguang Liu, Jiawei Wang, Tingwen Han, Cunguang Lou, Tie Liang, Hongrui Wang, Xiuling Liu
Summary: Intelligent prosthetic hand is an important branch of intelligent robotics that can remotely complete complex tasks and assist in rehabilitation training. This paper proposes a new multichannel fusion scheme to improve gesture recognition accuracy and enhances the performance of the network using an improved Temporal Convolutional Network.
APPLIED BIONICS AND BIOMECHANICS
(2022)
Article
Multidisciplinary Sciences
Patricio J. Cruz, Juan Pablo Vasconez, Ricardo Romero, Alex Chico, Marco E. Benalcazar, Robin Alvarez, Lorena Isabel Barona Lopez, Angel Leonardo Valdivieso Caraguay
Summary: This study proposes a reinforcement learning-based hand gesture recognition system that utilizes electromyography and inertial measurement unit signals. The system has potential in controlling video games, vehicles, and robots. By employing Deep Q-learning algorithm, an agent is created to classify the EMG-IMU signals and achieves an accuracy of 97.45%±1.02% for classification and 88.05%±3.10% for recognition, with an average observation time of 20 ms.
SCIENTIFIC REPORTS
(2023)
Article
Engineering, Electrical & Electronic
Francesco Santoni, Alessio De Angelis, Antonio Moschitta, Paolo Carbone
Summary: This article introduces a hand-tracking system based on magnetic positioning, which can detect the position and orientation of the hand using magnetic fields and calculate finger joint positions and hand positions using a kinematic model. The system can work in non-line-of-sight conditions and has potential for various applications.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Biotechnology & Applied Microbiology
Guangjie Yu, Ziting Deng, Zhenchen Bao, Yue Zhang, Bingwei He, Crescenzio Gallo, Gianluca Zaza
Summary: This study introduces a convolutional neural network-enhanced channel attention model for surface electromyography gesture recognition, and proposes preprocessing and voting window techniques to improve the accuracy and stability of the model.
BIOENGINEERING-BASEL
(2023)
Article
Computer Science, Artificial Intelligence
Ang Ke, Jian Huang, Jing Wang, Jiping He
Summary: This study provides a framework for robust hand grasp type classification during dynamic arm position changes by improving both the hardware and algorithm components. Experimental results showed that the classification accuracy of multi-modal EMG-FMG signals was increased by more than 10% compared with the EMG-only signal, and the proposed sequential decision algorithm improved the accuracy by more than 4% compared with other baseline models when using both EMG and FMG signals.
FRONTIERS IN NEUROROBOTICS
(2022)
Article
Robotics
Luis Vargas, He Huang, Yong Zhu, Xiaogang Hu
Summary: This study evaluated the recognition of object properties using myoelectric control and sensory feedback in controlling a prosthetic hand. The findings show that evoked haptic and proprioceptive feedback can facilitate closed-loop control of the prosthetic hand, allowing for simultaneous recognition of different object properties.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Multidisciplinary Sciences
Shiqi Wang, Kankan Wang, Tingping Yang, Yiming Li, Di Fan
Summary: This paper proposes an improved 3D-ResNet sign language recognition algorithm with enhanced hand features, aiming to improve the accuracy of sign language recognition. The algorithm detects the hand regions using the improved EfficientDet network and enhances the detection ability with dual channel and spatial attention modules. Additionally, an improved residual module is used to extract sign language features. Experimental results show that the proposed algorithm achieves higher recognition accuracy compared to other algorithms.
SCIENTIFIC REPORTS
(2022)
Review
Computer Science, Information Systems
Noraini Mohamed, Mumtaz Begum Mustafa, Nazean Jomhari
Summary: This paper reviewed the research progress of sign language in vision-based hand gesture recognition system from 2014 to 2020, indicating an active field with focus on data acquisition, data environment, and hand gesture representation. Recognition accuracy varies between signer dependent and signer independent studies, with a lack of progress in continuous gesture recognition, suggesting the need for further improvement towards a practical vision-based gesture recognition system.
Article
Computer Science, Information Systems
Xun Wang, Ke Sun, Ting Zhao, Wei Wang, Qing Gu
Summary: In this paper, a Dynamic Speed Warping (DSW) algorithm is proposed for one-shot learning of device-free acoustic gesture signals performed by different users. The design of DSW is based on the observation that gesture type is determined by the trajectory of hand components rather than the movement speed. By dynamically scaling the speed distribution and tracking the movement distance along the trajectory, DSW can effectively match gesture signals from different domains with a ten-fold difference in speeds. Experimental results show that DSW can achieve a recognition accuracy of 97% for gestures performed by unknown users while only using one training sample of each gesture type from four training users.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Cheng-Ying Yang, Yi-Nan Lin, Sheng-Kuan Wang, Victor R. L. Shen, Yi-Chih Tung, Frank H. C. Shen, Chun-Hsiang Huang
Summary: With the rapid development of IoT technology, various intelligent home appliances in the market are constantly innovating, and the public's requirements for residential safety and convenience are increasing. Aging societies are a challenge for countries around the world. Hand gesture recognition is gaining popularity in various fields, and creating a convenient and smart control system for home appliances for the elderly or disabled has become the objective of this study. The study uses Google MediaPipe to develop a hand tracking system and achieves high precision and recall values in recognizing hand gestures.
APPLIED ARTIFICIAL INTELLIGENCE
(2023)
Article
Engineering, Electrical & Electronic
Peiying Lin, Ruofan Zhuo, Shiyu Wang, Zhouyi Wu, Jiangtao Huangfu
Summary: This study proposes a gesture recognition method and device based on light sensing characteristics of LED screens, which eliminates the need for external sensors. By utilizing deep learning and optimized algorithms, the system achieves high accuracy in gesture recognition.
IEEE SENSORS JOURNAL
(2022)
Article
Multidisciplinary Sciences
Simon Tam, Mounir Boukadoum, Alexandre Campeau-Lecours, Benoit Gosselin
Summary: This paper introduces a real-time control strategy for myoelectric hand prostheses that emphasizes intuitive and responsive control using surface high-density electromyography and a convolutional neural network. The system achieves reliable gesture recognition in real-time testing, with positive predictive values exceeding 93% and low latency. The use of transfer learning significantly reduces setup time, making the system user-friendly and efficient.
SCIENTIFIC REPORTS
(2021)
Review
Engineering, Multidisciplinary
Ziming Chen, Huasong Min, Dong Wang, Ziwei Xia, Fuchun Sun, Bin Fang
Summary: Myoelectric control is crucial for prosthetic hands in rehabilitation. Current research focuses on developing myoelectric classifiers and control methods, but limits hand manipulation to simple tasks and neglects complex daily activities. This article reviews recent achievements in intention recognition and control strategy, analyzing challenges and opportunities for functionality-augmented prosthetic hands and user burden reduction.
Article
Automation & Control Systems
Si-Hwan Heo, Hyung-Soon Park
Summary: This article proposes a novel perception system called proximity perception-based grasping intelligence (P2GI) for achieving seamless control of a highly dexterous prosthetic hand. The system utilizes proximity sensors to map the point cloud of the object in real time, and a decision-making algorithm to infer the user's intended grasp posture. The system has been evaluated with ten subjects, showing high accuracy in grasp posture classification and task success rate.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2023)
Article
Neurosciences
Jianjun Meng, Zehan Wu, Songwei Li, Xiangyang Zhu
Summary: This study investigated the impact of gaze fixation and covert attention on motor imagery-based brain-computer interface (BCI) performance by designing a gaze fixation controlled experiment. Results showed that there was a significantly shorter gaze shift response time in congruent trials compared to incongruent trials. However, the lateralization index computed from the parietal and occipital areas did not correlate with BCI behavioral performance.
FRONTIERS IN HUMAN NEUROSCIENCE
(2022)
Article
Engineering, Biomedical
Chen Chen, Yang Yu, Xinjun Sheng, Jianjun Meng, Xiangyang Zhu
Summary: This study proposed a real-time hand gesture recognition method by decoding motor unit (MU) discharges across multiple motor tasks in a motion-wise way. The results showed that the proposed method is feasible and superior in motion tasks and gesture recognition, expanding the potential applications of neural decoding in human-machine interfaces.
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING
(2023)
Article
Engineering, Biomedical
Shiyong Su, Guohong Chai, Wei Xu, Jianjun Meng, Xinjun Sheng, Andre Mouraux, Xiangyang Zhu
Summary: The study investigates the electrophysiological mechanism underlying sensory feedback and multi-sensory integration in prosthetic control tasks. The results demonstrate that visual feedback plays a predominant role in grasping force control and box and block control tasks, while tactile feedback is essential for proprioceptive position perception. Tactile-visual integration is found when implemented congruently. The findings provide neural evidence for the functional mechanisms of sensory feedback in prosthetic control.
JOURNAL OF NEURAL ENGINEERING
(2023)
Article
Automation & Control Systems
Xingchen Yang, Jipeng Yan, Zongtian Yin, Honghai Liu
Summary: This article proposes a wearable ultrasound-based interface for simultaneous and proportional control of wrist rotation and hand grasp. A semisupervised learning framework is used to simplify the model calibration, and the proposed algorithms are verified through offline and online experiments. This study is the first to achieve online simultaneous and proportional control of wrist and hand kinematics using ultrasound and semisupervised learning.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2023)
Review
Multidisciplinary Sciences
Ning Jiang, Chen Chen, Jiayuan He, Jianjun Meng, Lizhi Pan, Shiyong Su, Xiangyang Zhu
Summary: A decade ago, researchers identified challenges in upper-limb prosthesis control and proposed four key technical challenges that could bridge the gap between academic and industrial state-of-the-art. This review examines the research efforts made in the past decade to address these challenges, as well as recent developments in deep learning methods, surface electromyogram decomposition, and open-source databases. The review also provides an outlook into the future of upper-limb prosthetic control research and development.
NATIONAL SCIENCE REVIEW
(2023)
Article
Neurosciences
Guangye Li, Shize Jiang, Jianjun Meng, Zehan Wu, Haiteng Jiang, Zhen Fan, Jie Hu, Xinjun Sheng, Dingguo Zhang, Gerwin Schalk, Liang Chen, Xiangyang Zhu
Summary: This study utilized intracranial electroencephalographic (iEEG) recordings from 36 human subjects to investigate the neural activation during hand movements in response to visual cues. The results revealed widespread neural activation across various brain regions, with specific focus on parietal, frontal, and occipital lobes. The findings also demonstrated temporal differences in neural activation and provided insights into the sensory and motor functions.
Article
Engineering, Biomedical
Yuxuan Wei, Xu Wang, Ruijie Luo, Ximing Mai, Songwei Li, Jianjun Meng
Summary: This study investigates the spatial-spectral representation of steady-state movement-related rhythm (SSMRR) in EEG during voluntary movements and assesses the effectiveness of decoding movement frequencies and limbs based on SSMRR. The results demonstrate that accurate movement frequencies and limbs can be decoded from EEG using spatial-spectral features extracted from SSMRR.
JOURNAL OF NEURAL ENGINEERING
(2023)
Article
Computer Science, Information Systems
Jikun Ai, Jianjun Meng, Ximing Mai, Xiangyang Zhu
Summary: Brain-computer interface (BCI) technology allows patients and humans to control a robotic arm, but current BCI systems struggle to accurately manipulate multi-degree robotic arms. This study proposes a novel BCI paradigm using moving flickering stimuli, attached to the robotic arm's gripper, to enhance control performance. Experimental results demonstrate that this new paradigm outperforms conventional fixed flickering stimuli in completing reaching and grasping tasks in an unstructured environment.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2023)
Article
Engineering, Biomedical
Chen Chen, Yang Yu, Xinjun Sheng, Jianjun Meng, Xiangyang Zhu
Summary: In this study, motor unit discharges were mapped to three degrees of freedom wrist movements using high-density surface electromyography (EMG). The proposed method efficiently estimated 3-DoF wrist torques, demonstrating its potential for advancing dexterous myoelectric control based on neural information.
IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING
(2023)
Article
Engineering, Biomedical
Ximing Mai, Jikun Ai, Minghao Ji, Xiangyang Zhu, Jianjun Meng
Summary: Brain-computer interfaces (BCIs) have emerged as a promising technique for individuals with motor disabilities to control external devices. However, one of the challenges is the long transition time when switching to a new target. This study proposed a hybrid BCI that combines steady-state visual evoked potential (SSVEP) and electrooculography (EOG), which significantly reduces the transition time and achieves continuous, fluent control.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2024)
Article
Engineering, Biomedical
Ximing Mai, Jikun Ai, Yuxuan Wei, Xiangyang Zhu, Jianjun Meng
Summary: This study proposes a novel augmentation method (PLTS) for SSVEP-BCI, which significantly improves the classification performance and ITR of SSVEP algorithms. This method enhances the practicality of SSVEP-based brain spellers under limited training data.
IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING
(2023)
Article
Engineering, Biomedical
Ximing Mai, Xinjun Sheng, Xiaokang Shu, Yidan Ding, Xiangyang Zhu, Jianjun Meng
Summary: The study proposes a novel calibration-free Bayesian approach for a brain-computer interface (BCI) by hybridizing SSVEP and electrooculography (EOG). The method successfully enables continuous control of external devices, with significantly improved accuracy and gaze-shifting time compared to existing approaches. This research provides a valuable framework for the development and application of plug-and-play BCIs in continuous control.
IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Zongtian Yin, Hanwei Chen, Xingchen Yang, Yifan Liu, Ning Zhang, Jianjun Meng, Honghai Liu
Summary: This study developed a miniaturized ultrasound device that could be integrated into a prosthetic hand socket to non-invasively detect muscle deformations. The experimental results demonstrated the efficacy of the designed prosthetic system based on the miniaturized A-mode ultrasound device, paving the way for an effective HMI system that could be widely used in prosthetic hand control.
INTELLIGENT ROBOTICS AND APPLICATIONS (ICIRA 2022), PT III
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Ximing Mai, Xinjun Sheng, Xiaokang Shu, Yidan Ding, Jianjun Meng, Xiangyang Zhu
Summary: Brain-computer interfaces can restore communication abilities for disabled individuals. A training-free hybrid BCI system was proposed to rapidly and accurately recognize new targets, improving upon the limitations of traditional systems.
INTELLIGENT ROBOTICS AND APPLICATIONS (ICIRA 2022), PT II
(2022)
