Cargando…
Proposing Metrics for Benchmarking Novel EEG Technologies Towards Real-World Measurements
Recent advances in electroencephalographic (EEG) acquisition allow for recordings using wet and dry sensors during whole-body motion. The large variety of commercially available EEG systems contrasts with the lack of established methods for objectively describing their performance during whole-body...
Autores principales: | , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Frontiers Media S.A.
2016
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4861738/ https://www.ncbi.nlm.nih.gov/pubmed/27242467 http://dx.doi.org/10.3389/fnhum.2016.00188 |
_version_ | 1782431245620215808 |
---|---|
author | Oliveira, Anderson S. Schlink, Bryan R. Hairston, W. David König, Peter Ferris, Daniel P. |
author_facet | Oliveira, Anderson S. Schlink, Bryan R. Hairston, W. David König, Peter Ferris, Daniel P. |
author_sort | Oliveira, Anderson S. |
collection | PubMed |
description | Recent advances in electroencephalographic (EEG) acquisition allow for recordings using wet and dry sensors during whole-body motion. The large variety of commercially available EEG systems contrasts with the lack of established methods for objectively describing their performance during whole-body motion. Therefore, the aim of this study was to introduce methods for benchmarking the suitability of new EEG technologies for that context. Subjects performed an auditory oddball task using three different EEG systems (Biosemi wet—BSM, Cognionics Wet—Cwet, Conionics Dry—Cdry). Nine subjects performed the oddball task while seated and walking on a treadmill. We calculated EEG epoch rejection rate, pre-stimulus noise (PSN), signal-to-noise ratio (SNR) and EEG amplitude variance across the P300 event window (CV(ERP)) from a subset of 12 channels common to all systems. We also calculated test-retest reliability and the subject’s level of comfort while using each system. Our results showed that using the traditional 75 μV rejection threshold BSM and Cwet epoch rejection rates are ~25% and ~47% in the seated and walking conditions respectively. However, this threshold rejects ~63% of epochs for Cdry in the seated condition and excludes 100% of epochs for the majority of subjects during walking. BSM showed predominantly no statistical differences between seated and walking condition for all metrics, whereas Cwet showed increases in PSN and CV(ERP), as well as reduced SNR in the walking condition. Data quality from Cdry in seated conditions were predominantly inferior in comparison to the wet systems. Test-retest reliability was mostly moderate/good for these variables, especially in seated conditions. In addition, subjects felt less discomfort and were motivated for longer recording periods while using wet EEG systems in comparison to the dry system. The proposed method was successful in identifying differences across systems that are mostly caused by motion-related artifacts and usability issues. We conclude that the extraction of the selected metrics from an auditory oddball paradigm may be used as a benchmark method for testing the performance of different EEG systems in mobile conditions. Moreover dry EEG systems may need substantial improvements to meet the quality standards of wet electrodes. |
format | Online Article Text |
id | pubmed-4861738 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2016 |
publisher | Frontiers Media S.A. |
record_format | MEDLINE/PubMed |
spelling | pubmed-48617382016-05-30 Proposing Metrics for Benchmarking Novel EEG Technologies Towards Real-World Measurements Oliveira, Anderson S. Schlink, Bryan R. Hairston, W. David König, Peter Ferris, Daniel P. Front Hum Neurosci Neuroscience Recent advances in electroencephalographic (EEG) acquisition allow for recordings using wet and dry sensors during whole-body motion. The large variety of commercially available EEG systems contrasts with the lack of established methods for objectively describing their performance during whole-body motion. Therefore, the aim of this study was to introduce methods for benchmarking the suitability of new EEG technologies for that context. Subjects performed an auditory oddball task using three different EEG systems (Biosemi wet—BSM, Cognionics Wet—Cwet, Conionics Dry—Cdry). Nine subjects performed the oddball task while seated and walking on a treadmill. We calculated EEG epoch rejection rate, pre-stimulus noise (PSN), signal-to-noise ratio (SNR) and EEG amplitude variance across the P300 event window (CV(ERP)) from a subset of 12 channels common to all systems. We also calculated test-retest reliability and the subject’s level of comfort while using each system. Our results showed that using the traditional 75 μV rejection threshold BSM and Cwet epoch rejection rates are ~25% and ~47% in the seated and walking conditions respectively. However, this threshold rejects ~63% of epochs for Cdry in the seated condition and excludes 100% of epochs for the majority of subjects during walking. BSM showed predominantly no statistical differences between seated and walking condition for all metrics, whereas Cwet showed increases in PSN and CV(ERP), as well as reduced SNR in the walking condition. Data quality from Cdry in seated conditions were predominantly inferior in comparison to the wet systems. Test-retest reliability was mostly moderate/good for these variables, especially in seated conditions. In addition, subjects felt less discomfort and were motivated for longer recording periods while using wet EEG systems in comparison to the dry system. The proposed method was successful in identifying differences across systems that are mostly caused by motion-related artifacts and usability issues. We conclude that the extraction of the selected metrics from an auditory oddball paradigm may be used as a benchmark method for testing the performance of different EEG systems in mobile conditions. Moreover dry EEG systems may need substantial improvements to meet the quality standards of wet electrodes. Frontiers Media S.A. 2016-05-10 /pmc/articles/PMC4861738/ /pubmed/27242467 http://dx.doi.org/10.3389/fnhum.2016.00188 Text en Copyright © 2016 Oliveira, Schlink, Hairston, König and Ferris. http://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution and reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. |
spellingShingle | Neuroscience Oliveira, Anderson S. Schlink, Bryan R. Hairston, W. David König, Peter Ferris, Daniel P. Proposing Metrics for Benchmarking Novel EEG Technologies Towards Real-World Measurements |
title | Proposing Metrics for Benchmarking Novel EEG Technologies Towards Real-World Measurements |
title_full | Proposing Metrics for Benchmarking Novel EEG Technologies Towards Real-World Measurements |
title_fullStr | Proposing Metrics for Benchmarking Novel EEG Technologies Towards Real-World Measurements |
title_full_unstemmed | Proposing Metrics for Benchmarking Novel EEG Technologies Towards Real-World Measurements |
title_short | Proposing Metrics for Benchmarking Novel EEG Technologies Towards Real-World Measurements |
title_sort | proposing metrics for benchmarking novel eeg technologies towards real-world measurements |
topic | Neuroscience |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4861738/ https://www.ncbi.nlm.nih.gov/pubmed/27242467 http://dx.doi.org/10.3389/fnhum.2016.00188 |
work_keys_str_mv | AT oliveiraandersons proposingmetricsforbenchmarkingnoveleegtechnologiestowardsrealworldmeasurements AT schlinkbryanr proposingmetricsforbenchmarkingnoveleegtechnologiestowardsrealworldmeasurements AT hairstonwdavid proposingmetricsforbenchmarkingnoveleegtechnologiestowardsrealworldmeasurements AT konigpeter proposingmetricsforbenchmarkingnoveleegtechnologiestowardsrealworldmeasurements AT ferrisdanielp proposingmetricsforbenchmarkingnoveleegtechnologiestowardsrealworldmeasurements |