1
|
Guerrero-Mendez CD, Lopez-Delis A, Blanco-Diaz CF, Bastos-Filho TF, Jaramillo-Isaza S, Ruiz-Olaya AF. Continuous reach-to-grasp motion recognition based on an extreme learning machine algorithm using sEMG signals. Phys Eng Sci Med 2024:10.1007/s13246-024-01454-5. [PMID: 38954380 DOI: 10.1007/s13246-024-01454-5] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/10/2023] [Accepted: 05/30/2024] [Indexed: 07/04/2024]
Abstract
Recognizing user intention in reach-to-grasp motions is a critical challenge in rehabilitation engineering. To address this, a Machine Learning (ML) algorithm based on the Extreme Learning Machine (ELM) was developed for identifying motor actions using surface Electromyography (sEMG) during continuous reach-to-grasp movements, involving multiple Degrees of Freedom (DoFs). This study explores feature extraction methods based on time domain and autoregressive models to evaluate ELM performance under different conditions. The experimental setup encompassed variations in neuron size, time windows, validation with each muscle, increase in the number of features, comparison with five conventional ML-based classifiers, inter-subjects variability, and temporal dynamic response. To evaluate the efficacy of the proposed ELM-based method, an openly available sEMG dataset containing data from 12 participants was used. Results highlight the method's performance, achieving Accuracy above 85%, F-score above 90%, Recall above 85%, Area Under the Curve of approximately 84% and compilation times (computational cost) of less than 1 ms. These metrics significantly outperform standard methods (p < 0.05). Additionally, specific trends were found in increasing and decreasing performance in identifying specific tasks, as well as variations in the continuous transitions in the temporal dynamics response. Thus, the ELM-based method effectively identifies continuous reach-to-grasp motions through myoelectric data. These findings hold promise for practical applications. The method's success prompts future research into implementing it for more reliable and effective Human-Machine Interface (HMI) control. This can revolutionize real-time upper limb rehabilitation, enabling natural and complex Activities of Daily Living (ADLs) like object manipulation. The robust results encourages further research and innovative solutions to improve people's quality of life through more effective interventions.
Collapse
Affiliation(s)
- Cristian D Guerrero-Mendez
- Faculty of Mechanical, Electronics and Biomedical Engineering, Antonio Nariño University (UAN), Bogota D.C, Colombia.
- Postgraduate Program in Electrical Engineering, Federal University of Espirito Santo (UFES), Vitoria, 29075-910, Brazil.
| | | | - Cristian F Blanco-Diaz
- Faculty of Mechanical, Electronics and Biomedical Engineering, Antonio Nariño University (UAN), Bogota D.C, Colombia
- Postgraduate Program in Electrical Engineering, Federal University of Espirito Santo (UFES), Vitoria, 29075-910, Brazil
| | - Teodiano F Bastos-Filho
- Postgraduate Program in Electrical Engineering, Federal University of Espirito Santo (UFES), Vitoria, 29075-910, Brazil
| | - Sebastian Jaramillo-Isaza
- Faculty of Mechanical, Electronics and Biomedical Engineering, Antonio Nariño University (UAN), Bogota D.C, Colombia
| | - Andres F Ruiz-Olaya
- Faculty of Mechanical, Electronics and Biomedical Engineering, Antonio Nariño University (UAN), Bogota D.C, Colombia
| |
Collapse
|
2
|
Jarque-Bou NJ, Gracia-Ibáñez V, Roda-Sales A, Bayarri-Porcar V, Sancho-Bru JL, Vergara M. Toward Early and Objective Hand Osteoarthritis Detection by Using EMG during Grasps. SENSORS (BASEL, SWITZERLAND) 2023; 23:2413. [PMID: 36904616 PMCID: PMC10006890 DOI: 10.3390/s23052413] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 01/19/2023] [Revised: 02/17/2023] [Accepted: 02/20/2023] [Indexed: 06/18/2023]
Abstract
The early and objective detection of hand pathologies is a field that still requires more research. One of the main signs of hand osteoarthritis (HOA) is joint degeneration, which causes loss of strength, among other symptoms. HOA is usually diagnosed with imaging and radiography, but the disease is in an advanced stage when HOA is observable by these methods. Some authors suggest that muscle tissue changes seem to occur before joint degeneration. We propose recording muscular activity to look for indicators of these changes that might help in early diagnosis. Muscular activity is often measured using electromyography (EMG), which consists of recording electrical muscle activity. The aim of this study is to study whether different EMG characteristics (zero crossing, wavelength, mean absolute value, muscle activity) via collection of forearm and hand EMG signals are feasible alternatives to the existing methods of detecting HOA patients' hand function. We used surface EMG to measure the electrical activity of the dominant hand's forearm muscles with 22 healthy subjects and 20 HOA patients performing maximum force during six representative grasp types (the most commonly used in ADLs). The EMG characteristics were used to identify discriminant functions to detect HOA. The results show that forearm muscles are significantly affected by HOA in EMG terms, with very high success rates (between 93.3% and 100%) in the discriminant analyses, which suggest that EMG can be used as a preliminary step towards confirmation with current HOA diagnostic techniques. Digit flexors during cylindrical grasp, thumb muscles during oblique palmar grasp, and wrist extensors and radial deviators during the intermediate power-precision grasp are good candidates to help detect HOA.
Collapse
Affiliation(s)
- Néstor J. Jarque-Bou
- Department of Mechanical Engineering and Construction, Universitat Jaume I, E12071 Castellón, Spain
| | | | | | | | | | | |
Collapse
|
3
|
Myoelectric Pattern Recognition Performance Enhancement Using Nonlinear Features. COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE 2022; 2022:6414664. [PMID: 35528339 PMCID: PMC9076314 DOI: 10.1155/2022/6414664] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 03/08/2022] [Revised: 04/07/2022] [Accepted: 04/11/2022] [Indexed: 11/17/2022]
Abstract
The multichannel electrode array used for electromyogram (EMG) pattern recognition provides good performance, but it has a high cost, is computationally expensive, and is inconvenient to wear. Therefore, researchers try to use as few channels as possible while maintaining improved pattern recognition performance. However, minimizing the number of channels affects the performance due to the least separable margin among the movements possessing weak signal strengths. To meet these challenges, two time-domain features based on nonlinear scaling, the log of the mean absolute value (LMAV) and the nonlinear scaled value (NSV), are proposed. In this study, we validate the proposed features on two datasets, the existing four feature extraction methods, variable window size, and various signal-to-noise ratios (SNR). In addition, we also propose a feature extraction method where the LMAV and NSV are grouped with the existing 11 time-domain features. The proposed feature extraction method enhances accuracy, sensitivity, specificity, precision, and F1 score by 1.00%, 5.01%, 0.55%, 4.71%, and 5.06% for dataset 1, and 1.18%, 5.90%, 0.66%, 5.63%, and 6.04% for dataset 2, respectively. Therefore, the experimental results strongly suggest the proposed feature extraction method, for taking a step forward with regard to improved myoelectric pattern recognition performance.
Collapse
|
4
|
Baygin M, Barua PD, Dogan S, Tuncer T, Key S, Acharya UR, Cheong KH. A Hand-Modeled Feature Extraction-Based Learning Network to Detect Grasps Using sEMG Signal. SENSORS (BASEL, SWITZERLAND) 2022; 22:2007. [PMID: 35271154 PMCID: PMC8914690 DOI: 10.3390/s22052007] [Citation(s) in RCA: 6] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 01/21/2022] [Revised: 02/11/2022] [Accepted: 02/21/2022] [Indexed: 06/14/2023]
Abstract
Recently, deep models have been very popular because they achieve excellent performance with many classification problems. Deep networks have high computational complexities and require specific hardware. To overcome this problem (without decreasing classification ability), a hand-modeled feature selection method is proposed in this paper. A new shape-based local feature extractor is presented which uses the geometric shape of the frustum. By using a frustum pattern, textural features are generated. Moreover, statistical features have been extracted in this model. Textures and statistics features are fused, and a hybrid feature extraction phase is obtained; these features are low-level. To generate high level features, tunable Q factor wavelet transform (TQWT) is used. The presented hybrid feature generator creates 154 feature vectors; hence, it is named Frustum154. In the multilevel feature creation phase, this model can select the appropriate feature vectors automatically and create the final feature vector by merging the appropriate feature vectors. Iterative neighborhood component analysis (INCA) chooses the best feature vector, and shallow classifiers are then used. Frustum154 has been tested on three basic hand-movement sEMG datasets. Hand-movement sEMG datasets are commonly used in biomedical engineering, but there are some problems in this area. The presented models generally required one dataset to achieve high classification ability. In this work, three sEMG datasets have been used to test the performance of Frustum154. The presented model is self-organized and selects the most informative subbands and features automatically. It achieved 98.89%, 94.94%, and 95.30% classification accuracies using shallow classifiers, indicating that Frustum154 can improve classification accuracy.
Collapse
Affiliation(s)
- Mehmet Baygin
- Department of Computer Engineering, College of Engineering, Ardahan University, Ardahan 75000, Turkey;
| | - Prabal Datta Barua
- School of Business (Information System), University of Southern Queensland, Toowoomba, QLD 4350, Australia;
- Faculty of Engineering and Information Technology, University of Technology Sydney, Sydney, NSW 2007, Australia
- Cogninet Brain Team, Cogninet Australia, Sydney, NSW 2010, Australia
| | - Sengul Dogan
- Department of Digital Forensics Engineering, College of Technology, Firat University, Elazig 23119, Turkey; (S.D.); (T.T.)
| | - Turker Tuncer
- Department of Digital Forensics Engineering, College of Technology, Firat University, Elazig 23119, Turkey; (S.D.); (T.T.)
| | - Sefa Key
- Department of Orthopedics and Traumatology, Bingöl State Hospital, Ministry of Health, Bingöl 12000, Turkey;
| | - U. Rajendra Acharya
- Department of Electronics and Computer Engineering, Ngee Ann Polytechnic, Singapore 599489, Singapore
- Department of Biomedical Engineering, School of Science and Technology, SUSS University, Singapore 599494, Singapore
- Department of Biomedical Informatics and Medical Engineering, Asia University, Taichung 41354, Taiwan
| | - Kang Hao Cheong
- Science, Mathematics and Technology Cluster, Singapore University of Technology and Design, Singapore 487372, Singapore
| |
Collapse
|
5
|
Fan J, Jiang M, Lin C, Li G, Fiaidhi J, Ma C, Wu W. Improving sEMG-based motion intention recognition for upper-limb amputees using transfer learning. Neural Comput Appl 2021. [DOI: 10.1007/s00521-021-06292-0] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/29/2022]
|