Can I read Motor Imagery Classification Combining Riemannian Geometry and Artificial Neural Networks on EtoBox?

Motor Imagery Classification Combining Riemannian Geometry and Artificial Neural Networks by Hubert Cecotti; Girish Tiwale is a book available to read on EtoBox.

What is Motor Imagery Classification Combining Riemannian Geometry and Artificial Neural Networks about?

based on non-invasive electroencephalography provide a means of communication for people with severe disabilities. BCI based on the detection of motor imagery can be used for both communication and rehabilitation purposes . For transferring BCIs outside of the lab to clinical settings, it is necessary to have a high accuracy. The current state of the art techniques includes the use of distance based on the Riemannian geometry. In this paper, we propose a new pattern recognition system for the multiclass classification of brain evoked responses corresponding to motor imagery. The method is based on the combination of features based on Riemannian geometry obtained from 15 frequency bands from 8 24 Hz to cover the mu and beta bands, and a feedforward neural network for the classification. We compare the performance of the multi-layer perceptron (MLP) and the extreme learning machine (ELM) classifiers. The system has been assessed on two publicly available datasets. The kappa value for 4-class is 0.53. The average binary classification across the six pairwise tasks is 80.83%. The results support the conclusion that multi-band classification can be successfully achieved using artificial

Author

Hubert Cecotti; Girish Tiwale

Publisher

Springer International Publishing

Published

2023

Language

EN

ISBN

9783031235986

Subjects

Technology, Science, Computer Science