Real-Time IMU-Based Kinematics in the Presence of Wireless Data Drop DOI
Kezhe Zhu, Dongxuan Li, Jinxuan Li

et al.

IEEE Journal of Biomedical and Health Informatics, Journal Year: 2024, Volume and Issue: 28(11), P. 6512 - 6524

Published: July 23, 2024

Wireless inertial motion capture holds promise for real-time human-machine interfaces and home-based rehabilitation applications. However, wireless data drop can cause significant estimation errors deteriorating performance or even making the system unusable. It is currently unclear how to estimate non-periodic kinematics with wearable measurement units (IMUs) in presence of (packet loss). We thus propose a novel inference encoder-decoder network model during dynamic movement. Twenty-four healthy subjects performed yoga, golf, swimming, dance, badminton movement activities while wearing IMUs 10-90% each IMU's were randomly removed determine effects on accuracy without proposed model. Results demonstrated reduction RMSE 45.2% 51.5% upper limb kinematic compared No Prediction strategy, 19.1% 31.3% an baseline LSTM In addition, has significantly less error (p<0.05) than strategy 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80% drop. These results could enable wearable, IMU analysis assessment reduced varying amounts further facilitate interaction medical treatment.

Language: Английский

A Review of Hand Gesture Recognition Systems Based on Noninvasive Wearable Sensors DOI Creative Commons
Rayane Tchantchane, Hao Zhou, Shen Zhang

et al.

Advanced Intelligent Systems, Journal Year: 2023, Volume and Issue: 5(10)

Published: July 20, 2023

Hand gesture, one of the essential ways for a human to convey information and express intuitive intention, has significant degree differentiation, substantial flexibility, high robustness transmission make hand gesture recognition (HGR) research hotspots in fields human–human human–computer or human–machine interactions. Noninvasive, on‐body sensors can monitor, track, recognize gestures various applications such as sign language recognition, rehabilitation, myoelectric control prosthetic hands interface (HMI), many other applications. This article systematically reviews recent achievements from noninvasive upper‐limb sensing techniques HGR, multimodal fusion gain additional user information, wearable algorithms obtain more reliable robust performance. Research challenges, progress, emerging opportunities sensor‐based HGR systems are also analyzed provide perspectives future progress.

Language: Английский

Citations

45
A Review of Myoelectric Control for Prosthetic Hand Manipulation DOI Creative Commons
Ziming Chen, Huasong Min, Dong Wang

et al.

Biomimetics, Journal Year: 2023, Volume and Issue: 8(3), P. 328 - 328

Published: July 24, 2023

Myoelectric control for prosthetic hands is an important topic in the field of rehabilitation. Intuitive and intelligent myoelectric can help amputees to regain upper limb function. However, current research efforts are primarily focused on developing rich classifiers biomimetic methods, limiting hand manipulation simple grasping releasing tasks, while rarely exploring complex daily tasks. In this article, we conduct a systematic review recent achievements two areas, namely, intention recognition strategy research. Specifically, focus advanced methods motion types, discrete classification, continuous estimation, unidirectional control, feedback shared control. addition, based above review, analyze challenges opportunities directions functionality-augmented user burden reduction, which overcome limitations provide development prospects future

Language: Английский

Citations

35
Non-invasive Techniques for Muscle Fatigue Monitoring: A Comprehensive Survey DOI Creative Commons
Na Li, Rui Zhou,

B. Sai Krishna

et al.

ACM Computing Surveys, Journal Year: 2024, Volume and Issue: 56(9), P. 1 - 40

Published: Feb. 20, 2024

Muscle fatigue represents a complex physiological and psychological phenomenon that impairs physical performance increases the risks of injury. It is important to continuously monitor levels for early detection management fatigue. The classification muscle also provide information in human-computer interactions (HMI), sports injuries performance, ergonomics, prosthetic control. With this purpose mind, review first provides an overview mechanisms its biomarkers further enumerates various non-invasive techniques commonly used monitoring literature, including electromyogram (EMG), which records electrical activity during contractions, mechanomyogram (MMG), vibration signals fibers, near-infrared spectroscopy (NIRS), measures amount oxygen muscle, ultrasound (US), deformation contractions. This introduces principle mechanism, parameters detection, application advantages disadvantages each technology detail. To conclude, limitations/challenges need be addressed future research area are presented.

Language: Английский

Citations

9
Hand Gesture Recognition from Surface Electromyography Signals with Graph Convolutional Network and Attention Mechanisms DOI
Hao Zhou, Hoang Thanh Le, Shen Zhang

et al.

IEEE Sensors Journal, Journal Year: 2025, Volume and Issue: 25(5), P. 9081 - 9092

Published: Jan. 15, 2025

Language: Английский

Citations

1
A Multipurpose Human–Machine Interface via 3D-Printed Pressure-Based Force Myography DOI
Hao Zhou, Charbel Tawk, Gürsel Alıcı

et al.

IEEE Transactions on Industrial Informatics, Journal Year: 2024, Volume and Issue: 20(6), P. 8838 - 8849

Published: March 29, 2024

Artificially intelligent (AI), powerful, and reliable human–machine interfaces (HMIs) are highly desired for wearable technologies, which proved to be the next advancement when it comes humans interacting with physical, digital, mixed environments. To demonstrate them, here we report on an innovative noninvasive, lightweight, low-cost, wearable, soft pressure-based force myography (pFMG) HMI in form of armband. The armband acquires stable mechanical biosignals air pressure information response forces induced by muscle activity consisting contraction relaxation that deform its pressure-sensitive chambers (PSCs). PSCs characterized a fast biosignal, negligible hysteresis, repeatability, reproducibility, reliability, stability, minimal calibration requirements, durability (more than 1 500 000 cycles). pFMG is resistant sweat, body hair present skin, worn cloth, scars, resilient external deformations. We capability versatility pFMG-based interact control collaborative robot manipulators, robotic prosthetic hands, drones, computer games, any system where loop. signals generated through implementation machine learning algorithm decode classify acquired different hand gestures rapidly accurately recognize intentions user. easy direct fabrication customization addition ability gesture reliably based makes ideal integrated into AI-powered applications.

Language: Английский

Citations

8
Hand Gesture Recognition Across Various Limb Positions Using a Multimodal Sensing System Based on Self-Adaptive Data-Fusion and Convolutional Neural Networks (CNNs) DOI
Shen Zhang, Hao Zhou, Rayane Tchantchane

et al.

IEEE Sensors Journal, Journal Year: 2024, Volume and Issue: 24(11), P. 18633 - 18645

Published: April 22, 2024

This study explores the challenge of hand gesture recognition across various limb positions using a new co-located multi-modal armband system incorporating Surface Electromyography (sEMG) and Pressure-based Force Myography (pFMG) sensors. Conventional Machine Learning (ML) algorithms Convolutional Neural Networks models (CNNs) were evaluated for accurately recognizing gestures. A comprehensive investigation was conducted, encompassing feature-level decision-level CNN models, alongside advanced fusion techniques to enhance performance. research consistently demonstrates superiority revealing their potential in extracting intricate patterns from raw sensor data. The showcased significant accuracy improvements over single-modality approaches, emphasizing synergistic effects sensing. Notably, achieved an 88.34% self-adaptive 87.79% fusion, outperforming Linear Discriminant Analysis (LDA) with 83.33% when considering all nine Furthermore, relationship between number gestures accuracy, high levels ranging 88% 100% 2-9 remarkable 98% commonly used five underscores adaptability CNNs effectively capturing complex complementation data varying positions, advancing field recognition, high-level data-fusion deep learning (DL) wearable sensing systems. provides valuable contributions into how sensor/data coupled ML methods, enhances ultimately paving way more effective adaptable technology applications.

Language: Английский

Citations

8
Recent Advances in Biomimetics for the Development of Bio-Inspired Prosthetic Limbs DOI Creative Commons

Pavitra Varaganti,

Soonmin Seo

Biomimetics, Journal Year: 2024, Volume and Issue: 9(5), P. 273 - 273

Published: April 30, 2024

Recent advancements in biomimetics have spurred significant innovations prosthetic limb development by leveraging the intricate designs and mechanisms found nature. Biomimetics, also known as “nature-inspired engineering”, involves studying emulating biological systems to address complex human challenges. This comprehensive review provides insights into latest trends biomimetic prosthetics, focusing on knowledge from natural biomechanics, sensory feedback mechanisms, control closely mimic appendages. Highlighted breakthroughs include integration of cutting-edge materials manufacturing techniques such 3D printing, facilitating seamless anatomical limbs. Additionally, incorporation neural interfaces enhances movement, while technologies like scanning enable personalized customization, optimizing comfort functionality for individual users. Ongoing research efforts hold promise further advancements, offering enhanced mobility individuals with loss or impairment. illuminates dynamic landscape technology, emphasizing its transformative potential rehabilitation assistive technologies. It envisions a future where solutions seamlessly integrate body, augmenting both quality life.

Language: Английский

Citations

7
Flexible wearable ionogels: Classification, fabrication, properties and applications DOI

Ying‐Ao Zhang,

Ke Ma, Kezheng Chen

et al.

Sensors and Actuators A Physical, Journal Year: 2024, Volume and Issue: 372, P. 115325 - 115325

Published: March 30, 2024

Language: Английский

Citations

5
IMU Shoulder Angle Estimation: Effects of Sensor-to-Segment Misalignment and Sensor Orientation Error DOI Creative Commons
Kezhe Zhu, Jinxuan Li, Dongxuan Li

et al.

IEEE Transactions on Neural Systems and Rehabilitation Engineering, Journal Year: 2023, Volume and Issue: 31, P. 4481 - 4491

Published: Jan. 1, 2023

Accurate shoulder joint angle estimation is crucial for analyzing kinematics and kinetics across a spectrum of movement applications including in athletic performance evaluation, injury prevention, rehabilitation. However, accurate IMU-based challenging the specific influence key error factors on unclear. We thus propose an analytical model based quaternions rotation vectors that decouples quantifies effects two factors, namely sensor-to-segment misalignment sensor orientation error, error. To validate this model, we conducted experiments involving twenty-five subjects who performed five activities: yoga, golf, swimming, dance, badminton. Results showed improving along segment's extension/flexion dimension had most significant impact reducing magnitude Specifically, 1° improvement thorax upper arm calibration resulted reduction 0.40° 0.57° magnitude. In comparison, IMU heading was only roughly half as effective (0.23° per 1°). This study clarifies relationship between its contributing identifies strategies these factors. These findings have implications enhancing accuracy estimation, thereby facilitating advancements limb rehabilitation, human-machine interaction, evaluation.

Language: Английский

Citations

10
Data Acquisition and Management in Industrial Businesses Fundamentals and Methodologies DOI

Xiaofeng Li

Journal of Enterprise and Business Intelligence, Journal Year: 2025, Volume and Issue: unknown, P. 030 - 039

Published: Jan. 5, 2025

In industrial enterprises, data acquisition is an essential procedure, basically in the industry 4.0 context. It entails taking signals and converting them into digital values that a computer can manipulate. order to transform analog waveforms modern for further processing, information gathering systems are essential. This article focuses on process of acquiring enterprises throughout age Industry reviewing constituents significance accurate dependable portraying processes. addition, study examines classification production according criteria influence accessibility, as well various techniques approaches used acquisition. The limitations human collection highlighted, along with benefits automated semi-automated capturing technologies. Management support may get from automation systems, which also investigated research. Using dedicated servers communications protocols consolidate data, it investigates issues industry-wide fragmentation systems. research goes deeper how machine vision, barcodes, RFID devices gather data. Finally, paper emphasizes need analyzing company's organizational technical environment proposes strategy building Manufacturing Information Acquisition System (MIAS).

Language: Английский

Citations

0