Format

Send to

Choose Destination
Methods. 2017 Oct 1;129:96-107. doi: 10.1016/j.ymeth.2017.06.019. Epub 2017 Jun 22.

A machine learning approach for automated wide-range frequency tagging analysis in embedded neuromonitoring systems.

Author information

1
Energy Efficient Embedded Systems (EEES) Lab - DEI, University of Bologna, Italy. Electronic address: fabio.montagna@unibo.it.
2
Center for Mind/Brain Sciences, University of Trento, Italy. Electronic address: marco.buiatti@unitn.it.
3
Energy Efficient Embedded Systems (EEES) Lab - DEI, University of Bologna, Italy. Electronic address: simone.benatti@unibo.it.
4
Energy Efficient Embedded Systems (EEES) Lab - DEI, University of Bologna, Italy. Electronic address: davide.rossi@unibo.it.
5
Energy Efficient Embedded Digital Architectures (E3DA) Unit - ICT Center, Fondazione Bruno Kessler, Italy. Electronic address: efarella@fbk.eu.
6
Energy Efficient Embedded Systems (EEES) Lab - DEI, University of Bologna, Italy; Integrated System Laboratory ETH, Zurich, Switzerland. Electronic address: luca.benini@unibo.it.

Abstract

EEG is a standard non-invasive technique used in neural disease diagnostics and neurosciences. Frequency-tagging is an increasingly popular experimental paradigm that efficiently tests brain function by measuring EEG responses to periodic stimulation. Recently, frequency-tagging paradigms have proven successful with low stimulation frequencies (0.5-6Hz), but the EEG signal is intrinsically noisy in this frequency range, requiring heavy signal processing and significant human intervention for response estimation. This limits the possibility to process the EEG on resource-constrained systems and to design smart EEG based devices for automated diagnostic. We propose an algorithm for artifact removal and automated detection of frequency tagging responses in a wide range of stimulation frequencies, which we test on a visual stimulation protocol. The algorithm is rooted on machine learning based pattern recognition techniques and it is tailored for a new generation parallel ultra low power processing platform (PULP), reaching performance of more that 90% accuracy in the frequency detection even for very low stimulation frequencies (<1Hz) with a power budget of 56mW.

KEYWORDS:

BCI; EEG; Embedded systems; Frequency-tagging; Machine learning; SVM

PMID:
28647609
DOI:
10.1016/j.ymeth.2017.06.019
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center