DocumentCode
2736415
Title
A neural network based classifier for the identification of simple finger motion
Author
Heinz, Michael ; Knapp, R. Benjamin
Author_Institution
Dept. of Electr. Eng., San Jose State Univ., CA, USA
Volume
3
fYear
1996
fDate
3-6 Jun 1996
Firstpage
1606
Abstract
The question of whether electromyographic (EMG) data from a single region of the forearm can be used to distinguish between various simple classes of finger motion is examined. Extensive clustering of data is performed to identify useful features for pattern classification. Sets of neural networks are trained to classify movements from each possible pairing of fingers. A multilayered network is constructed to distinguish between all five possible feature types
Keywords
biomechanics; data acquisition; electromyography; feature extraction; feedforward neural nets; medical computing; pattern classification; EMG data; data acquisition; data clustering; electromyographic data; feature selection; finger motion; forearm; multilayer neural network; muscle contraction; neural classifier; pattern classification; Biological control systems; Data acquisition; Electromyography; Fingers; Frequency; Medical signal detection; Neural networks; Position measurement; Sequential analysis; Signal analysis;
fLanguage
English
Publisher
ieee
Conference_Titel
Neural Networks, 1996., IEEE International Conference on
Conference_Location
Washington, DC
Print_ISBN
0-7803-3210-5
Type
conf
DOI
10.1109/ICNN.1996.549140
Filename
549140
Link To Document