Thumb-tip force estimation and analysis
Upper limb amputation that includes finger amputation is increasing every year due to accidents, diseases and congenital amputation. The loss of finger especially thumb could limit the proper hand functions and thus affect human daily activities. As a solution, a prosthetic thumb can be worn as a re...
Saved in:
| Main Author: | |
|---|---|
| Format: | Monograph |
| Language: | English |
| Published: |
s.n
2012
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/36538/1/EDW_A12-479-1270.pdf http://irep.iium.edu.my/36538/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Upper limb amputation that includes finger amputation is increasing every year due to accidents, diseases and congenital amputation. The loss of finger especially thumb could limit the proper hand functions and thus affect human daily activities. As a solution, a prosthetic thumb can be worn as a replacement to the real thumb. Natural controlled of the prosthetic device are desired and can be achieved by controlling the movement and force based on the real thumb model. The real thumb operates by muscles through muscles contraction. During contraction, muscle fibres inside the muscles are excited and electrical signals known as Electromyogram (EMG) signals are generated. These signals can be measured non-invasively using surface electrodes and can be used to control the prosthetic thumb. In this research, the EMG signals are measured simultaneously with the thumb-tip forces at different joint angles from five subjects. The muscles under consideration are the four thumb intrinsic muscles that are located at the outermost layer namely First Dorsal Interosseus (FDI), Flexor Pollicis Brevis (FPB), Abductor Pollicis Brevis (APB) and Adductor Pollicis (AP). The collected signals from the muscles are extracted in order to establish the relationship between the EMG signals and thumb-tip forces at different joint angles. The model of the relationships is developed by using Artificial Neural Network (ANN) with the EMG signals are set as the inputs and the thumb-tip force and joint angles are set as the output of the network. The performances of the established network are evaluated by calculating the root mean square error (RMSE) between the actual outputs and the estimated outputs. The network with the smallest RMSE could be used to control prosthetic thumb using EMG signals |
|---|