TY - GEN
T1 - Using Morphological-Linear Neural Network for Upper Limb Movement Intention Recognition from EEG Signals
AU - Hernández, Gerardo
AU - Hernández, Luis G.
AU - Zamora, Erik
AU - Sossa, Humberto
AU - Antelis, Javier M.
AU - Mendoza-Montoya, Omar
AU - Falcón, Luis E.
N1 - Publisher Copyright:
© 2019, Springer Nature Switzerland AG.
PY - 2019
Y1 - 2019
N2 - This study aims to compare classical and Morphological-Linear Neural Network (MLNN) algorithms for the intention recognition to perform different movements from electroencephalographic (EEG) signals. Three classification models were implemented and assessed to decode EEG motor imagery signals: (i) Morphological-Linear Neural Network (MLNN) (ii) Support Vector Machine (SVM) and (iii) Multilayer perceptron (MLP). Real EEG signals recorded during robot-assisted rehabilitation therapy were used to evaluate the performance of the proposed algorithms in the classification of three classes (relax, movement intention A Int A and movement intention B Int B) using multi-CSP based features extracted from EEG signals. The results of a ten-fold cross validation show similar results in terms of classification accuracy for the SVM and MLNN models. However, the number of parameters used in each model varies considerably (the MLNN model use less parameters than the SVM). This study indicates potential application of MLNNs for decoding movement intentions and its use to develop more natural and intuitive robot assisted neurorehabilitation therapies.
AB - This study aims to compare classical and Morphological-Linear Neural Network (MLNN) algorithms for the intention recognition to perform different movements from electroencephalographic (EEG) signals. Three classification models were implemented and assessed to decode EEG motor imagery signals: (i) Morphological-Linear Neural Network (MLNN) (ii) Support Vector Machine (SVM) and (iii) Multilayer perceptron (MLP). Real EEG signals recorded during robot-assisted rehabilitation therapy were used to evaluate the performance of the proposed algorithms in the classification of three classes (relax, movement intention A Int A and movement intention B Int B) using multi-CSP based features extracted from EEG signals. The results of a ten-fold cross validation show similar results in terms of classification accuracy for the SVM and MLNN models. However, the number of parameters used in each model varies considerably (the MLNN model use less parameters than the SVM). This study indicates potential application of MLNNs for decoding movement intentions and its use to develop more natural and intuitive robot assisted neurorehabilitation therapies.
KW - Brain-computer interfaces
KW - Electroencephalogram
KW - Machine learning
KW - Morphological-linear neural network
KW - Movement planing
UR - http://www.scopus.com/inward/record.url?scp=85068335278&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-21077-9_36
DO - 10.1007/978-3-030-21077-9_36
M3 - Contribución a la conferencia
AN - SCOPUS:85068335278
SN - 9783030210762
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 389
EP - 397
BT - Pattern Recognition - 11th Mexican Conference, MCPR 2019, Proceedings
A2 - Carrasco-Ochoa, Jesús Ariel
A2 - Martínez-Trinidad, José Francisco
A2 - Olvera-López, José Arturo
A2 - Salas, Joaquín
PB - Springer Verlag
T2 - 11th Mexican Conference on Pattern Recognition, MCPR 2019
Y2 - 26 June 2019 through 29 June 2019
ER -