Classification of Anticipatory Signals for Grasp and Release from Surface Electromyography
Surface electromyography (sEMG) is a technique for recording natural muscle activation signals, which can serve as control inputs for exoskeletons and prosthetic devices. Previous experiments have incorporated these signals using both classical and pattern-recognition control methods in order to act...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Published: |
MDPI AG
2018
|
| Online Access: | http://hdl.handle.net/1721.1/114488 https://orcid.org/0000-0003-3451-8046 https://orcid.org/0000-0003-1338-8107 https://orcid.org/0000-0002-0119-1617 |
| _version_ | 1826214684208922624 |
|---|---|
| author | Siu, Ho Chit Shah, Julie A Stirling, Leia A. |
| author2 | Institute for Medical Engineering and Science |
| author_facet | Institute for Medical Engineering and Science Siu, Ho Chit Shah, Julie A Stirling, Leia A. |
| author_sort | Siu, Ho Chit |
| collection | MIT |
| description | Surface electromyography (sEMG) is a technique for recording natural muscle activation signals, which can serve as control inputs for exoskeletons and prosthetic devices. Previous experiments have incorporated these signals using both classical and pattern-recognition control methods in order to actuate such devices. We used the results of an experiment incorporating grasp and release actions with object contact to develop an intent-recognition system based on Gaussian mixture models (GMM) and continuous-emission hidden Markov models (HMM) of sEMG data. We tested this system with data collected from 16 individuals using a forearm band with distributed sEMG sensors. The data contain trials with shifted band alignments to assess robustness to sensor placement. This study evaluated and found that pattern-recognition-based methods could classify transient anticipatory sEMG signals in the presence of shifted sensor placement and object contact. With the best-performing classifier, the effect of label lengths in the training data was also examined. A mean classification accuracy of 75.96% was achieved through a unigram HMM method with five mixture components. Classification accuracy on different sub-movements was found to be limited by the length of the shortest sub-movement, which means that shorter sub-movements within dynamic sequences require larger training sets to be classified correctly. This classification of user intent is a potential control mechanism for a dynamic grasping task involving user contact with external objects and noise. Further work is required to test its performance as part of an exoskeleton controller, which involves contact with actuated external surfaces. |
| first_indexed | 2024-09-23T16:09:24Z |
| format | Article |
| id | mit-1721.1/114488 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T16:09:24Z |
| publishDate | 2018 |
| publisher | MDPI AG |
| record_format | dspace |
| spelling | mit-1721.1/1144882022-10-02T06:44:48Z Classification of Anticipatory Signals for Grasp and Release from Surface Electromyography Siu, Ho Chit Shah, Julie A Stirling, Leia A. Institute for Medical Engineering and Science Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. Department of Aeronautics and Astronautics Siu, Ho Chit Shah, Julie A Stirling, Leia A. Surface electromyography (sEMG) is a technique for recording natural muscle activation signals, which can serve as control inputs for exoskeletons and prosthetic devices. Previous experiments have incorporated these signals using both classical and pattern-recognition control methods in order to actuate such devices. We used the results of an experiment incorporating grasp and release actions with object contact to develop an intent-recognition system based on Gaussian mixture models (GMM) and continuous-emission hidden Markov models (HMM) of sEMG data. We tested this system with data collected from 16 individuals using a forearm band with distributed sEMG sensors. The data contain trials with shifted band alignments to assess robustness to sensor placement. This study evaluated and found that pattern-recognition-based methods could classify transient anticipatory sEMG signals in the presence of shifted sensor placement and object contact. With the best-performing classifier, the effect of label lengths in the training data was also examined. A mean classification accuracy of 75.96% was achieved through a unigram HMM method with five mixture components. Classification accuracy on different sub-movements was found to be limited by the length of the shortest sub-movement, which means that shorter sub-movements within dynamic sequences require larger training sets to be classified correctly. This classification of user intent is a potential control mechanism for a dynamic grasping task involving user contact with external objects and noise. Further work is required to test its performance as part of an exoskeleton controller, which involves contact with actuated external surfaces. Massachusetts Institute of Technology (Jeptha and Emily V. Wade Award) 2018-03-30T19:44:32Z 2018-03-30T19:44:32Z 2016-10 2016-07 2018-03-02T16:10:47Z Article http://purl.org/eprint/type/JournalArticle 1424-8220 http://hdl.handle.net/1721.1/114488 Siu, Ho, Julie Shah, and Leia Stirling. “Classification of Anticipatory Signals for Grasp and Release from Surface Electromyography.” Sensors 16, no. 12 (October 25, 2016): 1782. © 2016 MDPI AG https://orcid.org/0000-0003-3451-8046 https://orcid.org/0000-0003-1338-8107 https://orcid.org/0000-0002-0119-1617 http://dx.doi.org/10.3390/S16111782 Sensors Attribution 4.0 International (CC BY 4.0) https://creativecommons.org/licenses/by/4.0/ application/pdf MDPI AG Diversity |
| spellingShingle | Siu, Ho Chit Shah, Julie A Stirling, Leia A. Classification of Anticipatory Signals for Grasp and Release from Surface Electromyography |
| title | Classification of Anticipatory Signals for Grasp and Release from Surface Electromyography |
| title_full | Classification of Anticipatory Signals for Grasp and Release from Surface Electromyography |
| title_fullStr | Classification of Anticipatory Signals for Grasp and Release from Surface Electromyography |
| title_full_unstemmed | Classification of Anticipatory Signals for Grasp and Release from Surface Electromyography |
| title_short | Classification of Anticipatory Signals for Grasp and Release from Surface Electromyography |
| title_sort | classification of anticipatory signals for grasp and release from surface electromyography |
| url | http://hdl.handle.net/1721.1/114488 https://orcid.org/0000-0003-3451-8046 https://orcid.org/0000-0003-1338-8107 https://orcid.org/0000-0002-0119-1617 |
| work_keys_str_mv | AT siuhochit classificationofanticipatorysignalsforgraspandreleasefromsurfaceelectromyography AT shahjuliea classificationofanticipatorysignalsforgraspandreleasefromsurfaceelectromyography AT stirlingleiaa classificationofanticipatorysignalsforgraspandreleasefromsurfaceelectromyography |