A Generalizable Brain-Computer Interface (BCI) Using Machine Learning for Feature Discovery

Ewan S Nurse; Philippa J Karoly; David B Grayden; Dean R Freestone

doi:10.1371/journal.pone.0131328

A Generalizable Brain-Computer Interface (BCI) Using Machine Learning for Feature Discovery

PLoS One. 2015 Jun 26;10(6):e0131328. doi: 10.1371/journal.pone.0131328. eCollection 2015.

Authors

Ewan S Nurse¹, Philippa J Karoly², David B Grayden¹, Dean R Freestone³

Affiliations

¹ NeuroEngineering Laboratory, Department of Electrical & Electronic Engineering, The University of Melbourne, Melbourne, VIC, Australia, 3010; Centre for Neural Engineering, The University of Melbourne, Melbourne, VIC, Australia, 3010.
² NeuroEngineering Laboratory, Department of Electrical & Electronic Engineering, The University of Melbourne, Melbourne, VIC, Australia, 3010; Centre for Neural Engineering, The University of Melbourne, Melbourne, VIC, Australia, 3010; Department of Medicine St. Vincent's Hospital Melbourne, The University of Melbourne, Fitzroy, VIC, Australia, 3065.
³ NeuroEngineering Laboratory, Department of Electrical & Electronic Engineering, The University of Melbourne, Melbourne, VIC, Australia, 3010; Centre for Neural Engineering, The University of Melbourne, Melbourne, VIC, Australia, 3010; Department of Medicine St. Vincent's Hospital Melbourne, The University of Melbourne, Fitzroy, VIC, Australia, 3065; Department of Statistics, Columbia University, New York, New York, USA, 10027.

Abstract

This work describes a generalized method for classifying motor-related neural signals for a brain-computer interface (BCI), based on a stochastic machine learning method. The method differs from the various feature extraction and selection techniques employed in many other BCI systems. The classifier does not use extensive a-priori information, resulting in reduced reliance on highly specific domain knowledge. Instead of pre-defining features, the time-domain signal is input to a population of multi-layer perceptrons (MLPs) in order to perform a stochastic search for the best structure. The results showed that the average performance of the new algorithm outperformed other published methods using the Berlin BCI IV (2008) competition dataset and was comparable to the best results in the Berlin BCI II (2002-3) competition dataset. The new method was also applied to electroencephalography (EEG) data recorded from five subjects undertaking a hand squeeze task and demonstrated high levels of accuracy with a mean classification accuracy of 78.9% after five-fold cross-validation. Our new approach has been shown to give accurate results across different motor tasks and signal types as well as between subjects.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Brain-Computer Interfaces*
Humans
Machine Learning*
Neural Networks, Computer*

Grants and funding

This research was supported by the Victorian Life Sciences Computation Initiative (VLSCI), an initiative of the Victorian Government hosted by the University of Melbourne, Australia. Dr. Freestone acknowledges the support of the Australian-American Fulbright Commission. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.