Transformers in biosignal analysis: A review

Information Fusion, Journal Year: 2024, Volume and Issue: 114, P. 102697 - 102697

Published: Sept. 16, 2024

Language: Английский

---

A survey on hand gesture recognition based on surface electromyography: Fundamentals, methods, applications, challenges and future trends

Applied Soft Computing, Journal Year: 2024, Volume and Issue: 166, P. 112235 - 112235

Published: Sept. 11, 2024

Language: Английский

---

Estimating gait parameters from sEMG signals using machine learning techniques under different power capacity of muscle

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: April 12, 2025

Language: Английский

---

Real-time torque and joint angle estimation using electromyography signals and an LSTM deep learning model on edge computing platform

Iran Journal of Computer Science, Journal Year: 2025, Volume and Issue: unknown

Published: April 28, 2025

Language: Английский

---

Wearable high-density EMG sleeve for complex hand gesture classification and continuous joint angle estimation

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Aug. 9, 2024

Abstract High-density electromyography (HD-EMG) can provide a natural interface to enhance human–computer interaction (HCI). This study aims demonstrate the capability of novel HD-EMG forearm sleeve equipped with up 150 electrodes capture high-resolution muscle activity, decode complex hand gestures, and estimate continuous position via joint angle predictions. Ten able-bodied participants performed 37 movements grasps while EMG was recorded using sleeve. Simultaneously, an 18-sensor motion glove calculated 23 angles from fingers across all for training regression models. For classifying our decoding algorithm able differentiate between sequential $$97.3 \pm 0.3\%$$ 97.3±0.3% accuracy on 100 ms bin-by-bin basis. In separate mixed dataset consisting 19 randomly interspersed, performance achieved average bin-wise $$92.8 0.8\%$$ xmlns:mml="http://www.w3.org/1998/Math/MathML">92.8±0.8% . When evaluating decoders use in real-time scenarios, we found that reliably both movement transitions, achieving $$93.3 0.9\%$$ xmlns:mml="http://www.w3.org/1998/Math/MathML">93.3±0.9% set $$88.5 xmlns:mml="http://www.w3.org/1998/Math/MathML">88.5±0.9% set. Furthermore, estimated data, $$R^2$$ xmlns:mml="http://www.w3.org/1998/Math/MathML">R2 $$0.884 0.003$$ xmlns:mml="http://www.w3.org/1998/Math/MathML">0.884±0.003 $$0.750 0.008$$ xmlns:mml="http://www.w3.org/1998/Math/MathML">0.750±0.008 Median absolute error (MAE) kept below 10° joints, grand MAE $$1.8 0.04^\circ$$ xmlns:mml="http://www.w3.org/1998/Math/MathML">1.8±0.04∘ $$3.4 0.07^\circ$$ xmlns:mml="http://www.w3.org/1998/Math/MathML">3.4±0.07∘ datasets, respectively. We also assessed two modifications address specific challenges EMG-driven HCI applications. To minimize decoder latency, used method accounts reaction time by dynamically shifting cue labels time. reduce requirements, show pretraining models historical data provided increase compared were not pretrained when reducing in-session only one attempt each movement. The sleeve, combined sophisticated machine learning algorithms, be powerful tool gesture recognition estimation. technology holds significant promise applications HCI, such as prosthetics, assistive technology, rehabilitation, human–robot collaboration.

Language: Английский

---

A Novel TCN-LSTM Hybrid Model for sEMG-Based Continuous Estimation of Wrist Joint Angles

Sensors, Journal Year: 2024, Volume and Issue: 24(17), P. 5631 - 5631

Published: Aug. 30, 2024

Surface electromyography (sEMG) offers a novel method in human–machine interactions (HMIs) since it is distinct physiological electrical signal that conceals human movement intention and muscle information. Unfortunately, the nonlinear non-smooth features of sEMG signals often make joint angle estimation difficult. This paper proposes prediction model for continuous wrist motion changes based on signals. The proposed combines temporal convolutional network (TCN) with long short-term memory (LSTM) network, where TCN can sense local information mine deeper signals, while LSTM, its excellent capability, up lack ability to capture long-term dependence resulting better prediction. We validated publicly available Ninapro DB1 dataset by selecting first eight subjects picking three types wrist-dependent movements: flexion (WF), ulnar deviation (WUD), extension closed hand (WECH). Finally, TCN-LSTM was compared LSTM models. outperformed models terms root mean square error (RMSE) average coefficient determination (R2). achieved an RMSE 0.064, representing 41% reduction 52% model. also R2 0.93, indicating 11% improvement over 18%

Language: Английский

---

Analysis of a New Toothbrushing Technique through Plaque Removal Success

Open Journal of Stomatology, Journal Year: 2024, Volume and Issue: 14(03), P. 133 - 152

Published: Jan. 1, 2024

Background/Aims: Determining the levels of oral health and quality dental care are fundamental to building concepts health. This study aims assess toothbrushing techniques using a technical physical model, clarifying how children pre-adults learn brush their teeth. Materials Methods: Data were recorded from 23 participants, both male female various ages, proposed electronic toothbrush equipped with X-Y-Z axes pathways. The data, collected before after training experiments, processed MATLAB generate plots for three axes. Results: revealed that most parameter values, such as Mean Difference Between Amplitudes (MAV, 6.00), Wilson Amplitude (WAMP, 179.419), Average Coupling (AAC, 1.270), decreased experiments. Furthermore, average overall epoch lengths (AVG) showed 75% reduction in movement amplitude between two Conclusion: Dentist observations indicated which brushing methods acceptable or not. Analytical values suggest individuals technique effectively, medical clearly demonstrate success method.

Language: Английский

---

Evaluation of Approaches for Early Stroke Detection and Diagnosis Using EMG Data: Features, Techniques, and Challenges

International Journal of Intelligent Information Systems, Journal Year: 2024, Volume and Issue: 13(2), P. 29 - 42

Published: April 17, 2024

This paper provides a thorough analysis of the use electromyography (EMG) data in early stroke diagnosis and detection. Stroke continues to be major global cause disability death, which emphasises critical need for an accurate made quickly improve patient outcomes. Early detection is still difficult achieve, even with improvements medical imaging testing technologies. By detecting minute variations muscle activity linked symptoms, EMG offers viable method identification. The review delves into diverse methodologies strategies utilised leverage purpose diagnosis, encompassing application deep learning models machine algorithms. proposes structured framework classifying approaches using data, providing systematic way categorize compare different methodologies. concludes by highlighting revolutionary potential EMG-based techniques improving strokes earlier urging more study address current issues make clinical easier.

Language: Английский

---

Artificial intelligence on biomedical signals: technologies, applications, and future directions

Med-X, Journal Year: 2024, Volume and Issue: 2(1)

Published: Dec. 10, 2024

Language: Английский

---

EMG feature extraction and muscle selection for continuous upper limb movement regression

Biomedical Signal Processing and Control, Journal Year: 2024, Volume and Issue: 103, P. 107323 - 107323

Published: Dec. 24, 2024

Language: Английский

---