Myo-to-Gesture: Evolving AI-Based Innovations in Myoelectric Signal Processing – State-of-the-Art Survey and Research Proposal

Lecture notes in computer science, Journal Year: 2025, Volume and Issue: unknown, P. 241 - 252

Published: Jan. 1, 2025

Language: Английский

---

A Robotic Prosthetic Hand for Computer Mouse Operations

Advanced Intelligent Systems, Journal Year: 2025, Volume and Issue: unknown

Published: March 2, 2025

Individuals with upper limb amputation face significant difficulty in operating standard computer peripherals their prostheses, particularly the mouse, due to constraints imposed by current prosthetic hand designs and irregular shape of mouse. This work introduces a soft myoelectric prosthesis system for different physical mouses. The proposed has two tendon‐driven fingers three fin‐ray fingers. It enables an enveloping grasp mouse facilitates diverse finger‐based clicking scrolling operations. In addition, mapping between movements (myoelectric signals) operations is created combining neural network classifier post‐processing algorithm. Herein, system's performance evaluated through series cursor control tasks involving both able‐bodied amputee participants. experimental results, it shown that effectively individuals transradial operate using residual limbs. Compared traditional method based on surface electromyography, participants exhibit improved task (shorter completion time higher throughput) user experience (reduced muscle use questionnaire scores).

Language: Английский

---

Hardware and Software Design and Implementation of Surface-EMG-Based Gesture Recognition and Control System

Electronics, Journal Year: 2024, Volume and Issue: 13(2), P. 454 - 454

Published: Jan. 22, 2024

The continuous advancement of electronic technology has led to the gradual integration automated intelligent devices into various aspects human life. Motion gesture-based human–computer interaction systems offer abundant information, user-friendly functionalities, and visual cues. Surface electromyography (sEMG) signals enable decoding muscle movements, facilitating realization corresponding control functions. Considering inherent instability minuscule nature sEMG signals, this thesis proposes a dynamic time regularization algorithm enhance gesture recognition detection accuracy real-time system performance. application warping allows fusion three enabling for calculation similarity between sample model. This process facilitates ensures effective communication individuals 3D printed prosthesis. Utilizing algorithm, best feature model was generated by amalgamating six types classification A total 600 training evaluation experiments were performed, with each movement recognized 100 times. experimental tests demonstrate that prosthetic limb using temporal achieves an impressive 93.75%, surpassing performance traditional threshold switch.

Language: Английский

---

A noise-tolerant dual multi-classifier system with fuzzy model applied to the sEMG-based control of a bionic upper limb prosthesis

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 104, P. 107441 - 107441

Published: Jan. 10, 2025

Language: Английский

---

Gaussian process latent variable models-ANN based method for automatic features selection and dimensionality reduction for control of EMG-driven systems

Frontiers in Artificial Intelligence, Journal Year: 2025, Volume and Issue: 8

Published: Jan. 22, 2025

Electromyography (EMG) signals have gained significant attention due to their potential applications in prosthetics, rehabilitation, and human-computer interfaces. However, the dimensionality of EMG signal features poses challenges achieving accurate classification reducing computational complexity. To overcome such issues, this paper proposes a novel approach that integrates feature reduction techniques with an artificial neural network (ANN) classifier enhance accuracy high-dimensional classification. This aims improve while substantially costs, offering valuable implications for all EMG-related processes on data. The proposed methodology involves extracting time frequency domain from twelve channels signals, followed by using as PCA, LDA, PPCA, Lasso GPLVM, ANN. Our investigation revealed LDA is not appropriate dataset. models did any effect accuracy, but cost decreased significantly. In individual comparisons, GPLVM had shortest (29 s), which was significantly less than other ( p < 0.05), PCA following at approximately 35 s Relief 57 s, PPCA took 69 exhibited higher costs lower original set. Using best-performing features, possible sets 2, 3, 4 5 were tested, 5-feature set best performance. research demonstrates effectiveness selection improving movement recognition myoelectric control.

Language: Английский

---

“The Foot Can Do It”: Controlling the “Persistence” Prosthetic Arm Using the “Infinity-2” Foot Controller

Technologies, Journal Year: 2025, Volume and Issue: 13(3), P. 98 - 98

Published: March 1, 2025

The “Infinity” foot controller for controlling prosthetic arms has been improved in this paper several ways, including a sleeve that enables barefoot use, an sensor-controller unit design, and more intuitive control scheme allows gradual of finger actuation. Furthermore, the “Persistence Arm”, novel transradial arm prototype, is introduced. This below-the-elbow direct-drive wrist actuation system, thumb design with two degrees freedom, carbon fiber tendons actuating four forefingers. manufactured prototype underwent various tests to verify their efficacy. Wireless transmission speed showed maximum time delay less than 165 ms, giving almost instantaneous response from any user’s signal. Gripping quantified grip pulling forces as 2.8 12.7 kg, respectively. successfully gripped household items different shapes, weights, sizes. These results highlight potential alternative method possibility new 3D-printed designs replace costly prostheses market, which could potentially reduce high rejection rates upper limb prostheses.

Language: Английский

---

Exploring the import of mechatronics engineering in medicine: a review

Beni-Suef University Journal of Basic and Applied Sciences, Journal Year: 2025, Volume and Issue: 14(1)

Published: March 23, 2025

Abstract Background The interdisciplinary nature of mechatronics has spurred huge progress in medicine to facilitate the creation robotic surgery, wearable health monitoring, and bio-inspired robots. All these technologies enhance precision boost diagnostic capability, enable real-time patient monitoring. For example, robotic-assisted surgeries have recorded a 50% cut complications 40% reduction healing times, while technology enhanced early anomaly detection by 80%, saving emergency hospitalisation. Main body This review critically examines evolution applications focusing on problems including financial burdens, confidentiality data, compliance with regulation. Emphasis is placed heavily regulatory approval processes required organisations such as US Food Drug Administration (FDA) International Organisation for Standardisation (ISO) that typically delay use life-saving equipment 3–5 years. In addition, expensive price surgery systems (~$2 million per unit) extensive training (20–40 procedures be proficient) are inhibiting factors. New trends robots nanomedicine also considered here, which exhibited fantastic potential minimally invasive therapy, nanorobot-based cancer therapies tumour growth inhibition limiting systemic side effects. Conclusions To propel ethical sustainable adoption healthcare, this proposed development partnerships among engineers, clinicians, policymakers, simplifies clearance processes, designs low-cost, scalable products. Through avenues, can proceed revolutionise enhancing outcomes expanding accessibility cutting-edge medical technology.

Language: Английский

---

Spatial identification of manipulable objects for a bionic hand prosthesis

Applied Computer Science, Journal Year: 2025, Volume and Issue: 21(1), P. 18 - 30

Published: March 31, 2025

This article presents a method for the spatial identification of objects bionic upper limb prostheses, utilizing analysis digital images captured by an optoelectronic module based on ESP32-CAM and classified using neural network algorithms, specifically FOMO (MobileNetV2). Modern prostheses that imitate natural functions, as well their advantages significance restoring functionality human body, are analysed. An algorithm grip-type recognition system is proposed, integrating object shapes with myographic signals to enable accurate selection execution appropriate manipulations. The was trained set basic (spherical, rectangular, cylindrical), which achieved average accuracy over 89% processing time one image 2 ms. Due its compactness low cost, developed suitable integration into low-cost ensuring adaptation movements artificial shape manipulation minimizing risk slipping objects. proposed approach helps increase movement reduce dependence expensive complex technologies. has potential further improvement, it can operate handle scenarios involving multiple within camera's field view simultaneously.

Language: Английский

---

Design and In Vivo Evaluation of an Intraocular Electrode for Ciliary Muscle Biopotential Measurement in a Non-Human Primate Model of Human Accommodation

Biosensors, Journal Year: 2025, Volume and Issue: 15(4), P. 247 - 247

Published: April 13, 2025

The measurement of electrical potentials in the human body is becoming increasingly important healthcare as a valuable diagnostic parameter. In ophthalmology, while these signals are primarily used to assess retinal function, other applications, such recording accommodation-related biopotentials from ciliary muscle, remain poorly understood. Here, we present development and evaluation novel implantable ring electrode for muscle. Inspired by capsular tension rings, was fabricated using laser cutting, wiring, physical vapor deposition coating. constant impedance weight over simulated aging period 391 days, demonstrated electrode’s stability. vivo testing non-human primates further validated surgical handling long-term stability, with no delamination or tissue ingrowth after 100 days implantation. Recorded muscle (up 700 µV) exceeded amplitudes reported literature. While results promising, research needed investigate signal quality origin well correlation between activity. Ultimately, this will be an implanted device record control artificial lens designed restore accommodation individuals presbyopia.

Language: Английский

---

Investigating the Feasibility and Safety of Osseointegration With Neural Interfaces for Advanced Prosthetic Control

Cureus, Journal Year: 2025, Volume and Issue: unknown

Published: April 19, 2025

Osseointegrated neural interfaces (ONI), particularly in conjunction with peripheral nerve (PNIs), have emerged as a promising advancement for intuitive neuroprosthetics. PNIs can decode signals and allow precise prosthetic movement control bidirectional communication haptic feedback, while osseointegration address limitations of traditional socket-based prosthetics, such poor stability, limited dexterity, lack sensory feedback. This review explores advancements ONIs, including screw-fit press-fit systems their integration communication. ONIs integrated (OIPNIs) shown improvements signal fidelity, motor control, feedback compared to popular surface electromyography (sEMG) systems. Additionally, emerging technologies hybrid electrode designs (e.g., cuff sieve (CASE)) regenerative (RPNIs) show potential enhancing selectivity reducing complications micromotion scarring. Despite these advances, challenges remain, infection risk, degradation, variability long-term stability. Osseointegration combined advanced represents transformative approach offering more natural Further research is needed biocompatibility, reduce surgical invasiveness, explore machine learning personalized ONI designs. The findings this underscore the enhance embodiment quality life amputees highlight current pitfalls possible areas improvement future research.

Language: Английский

---