TY - GEN
T1 - A Novel Channel Selection Method Based on Partial KL Information Measure for EMG-based Motion Classification
AU - Shibanoki, T.
AU - Shima, K.
AU - Tsuji, T.
AU - Otsuka, A.
AU - Chin, T.
PY - 2009
Y1 - 2009
N2 - To control machines using electromyograms (EMGs), subjects' intentions have to be correctly estimated and classified. However, the accuracy of classification is greatly influenced by individual physical abilities and measuring positions, making it necessary to select optimal channel positions for each subject. This paper proposes a novel online channel selection method using probabilistic neural networks (PNNs). In this method, measured data are regarded as probability variables, and data dimensions are evaluated by a partial KL information measure that is newly defined as a metric of effective dimensions. In the experiments, channels were selected using this method, and EMGs measured from the forearm were classified. The results showed that the number of channels is reduced with 33.33 ± 11.8%, and the average classification rate using the selected channels is almost the same as that using all channels. This demonstrates that the method is capable of selecting effective channels for classification.
AB - To control machines using electromyograms (EMGs), subjects' intentions have to be correctly estimated and classified. However, the accuracy of classification is greatly influenced by individual physical abilities and measuring positions, making it necessary to select optimal channel positions for each subject. This paper proposes a novel online channel selection method using probabilistic neural networks (PNNs). In this method, measured data are regarded as probability variables, and data dimensions are evaluated by a partial KL information measure that is newly defined as a metric of effective dimensions. In the experiments, channels were selected using this method, and EMGs measured from the forearm were classified. The results showed that the number of channels is reduced with 33.33 ± 11.8%, and the average classification rate using the selected channels is almost the same as that using all channels. This demonstrates that the method is capable of selecting effective channels for classification.
KW - electromyogram
KW - Kullback-Leibler information
KW - partial Wilks' lambda
KW - pattern classification
KW - variable selection method
UR - http://www.scopus.com/inward/record.url?scp=84891495772&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84891495772&partnerID=8YFLogxK
U2 - 10.1007/978-3-540-92841-6_170
DO - 10.1007/978-3-540-92841-6_170
M3 - Conference contribution
AN - SCOPUS:84891495772
SN - 9783540928409
T3 - IFMBE Proceedings
SP - 694
EP - 698
BT - 13th International Conference on Biomedical Engineering - ICBME 2008
T2 - 13th International Conference on Biomedical Engineering, ICBME 2008
Y2 - 3 December 2008 through 6 December 2008
ER -