Intracerebral EEG Artifact Identification Using Convolutional
Neural Networks

J 2019

Intracerebral EEG Artifact Identification Using Convolutional Neural Networks

NEJEDLY, Petr, Jan CIMBALNIK, Petr KLIMES, Filip PLESINGER, Josef HALAMEK et. al.

Basic information

Original name

Intracerebral EEG Artifact Identification Using Convolutional Neural Networks

Authors

NEJEDLY, Petr (203 Czech Republic, guarantor), Jan CIMBALNIK (203 Czech Republic), Petr KLIMES (203 Czech Republic), Filip PLESINGER (203 Czech Republic), Josef HALAMEK (203 Czech Republic), Vaclav KREMEN (203 Czech Republic), Ivo VISCOR (203 Czech Republic), Benjamin H. BRINKMANN (840 United States of America), Martin PAIL (203 Czech Republic, belonging to the institution), Milan BRÁZDIL (203 Czech Republic, belonging to the institution), Gregory WORRELL (840 United States of America) and Pavel JURAK (203 Czech Republic)

Edition

NEUROINFORMATICS, TOTOWA, HUMANA PRESS INC, 2019, 1539-2791

Other information

Language

English

Type of outcome

Článek v odborném periodiku

Field of Study

30103 Neurosciences

Country of publisher

United States of America

Confidentiality degree

není předmětem státního či obchodního tajemství

References:

URL

Impact factor

Impact factor: 3.300

RIV identification code

RIV/00216224:14110/19:00107403

Organization unit

Faculty of Medicine

DOI

http://dx.doi.org/10.1007/s12021-018-9397-6

UT WoS

000464856900005

Keywords in English

Intracranial EEG (iEEG); Noise detection; Convolutional neural networks (CNN); Artifact probability matrix (APM)

Abstract

V originále

Manual and semi-automatic identification of artifacts and unwanted physiological signals in large intracerebral electroencephalographic (iEEG) recordings is time consuming and inaccurate. To date, unsupervised methods to accurately detect iEEG artifacts are not available. This study introduces a novel machine-learning approach for detection of artifacts in iEEG signals in clinically controlled conditions using convolutional neural networks (CNN) and benchmarks the method's performance against expert annotations. The method was trained and tested on data obtained from St Anne's University Hospital (Brno, Czech Republic) and validated on data from Mayo Clinic (Rochester, Minnesota, U.S.A). We show that the proposed technique can be used as a generalized model for iEEG artifact detection. Moreover, a transfer learning process might be used for retraining of the generalized version to form a data-specific model. The generalized model can be efficiently retrained for use with different EEG acquisition systems and noise environments. The generalized and specialized model F1 scores on the testing dataset were 0.81 and 0.96, respectively. The CNN model provides faster, more objective, and more reproducible iEEG artifact detection compared to manual approaches.

Links

GAP103/11/0933, research and development project

Name: Analýza vysokofrekvenčního EEG signálu z hlubokých mozkových elektrod

Investor: Czech Science Foundation

NV16-33798A, research and development project

Name: Modulace funkční konektivity kortikálních sítí vlivem STN DBS

Citovat

NEJEDLY, Petr, Jan CIMBALNIK, Petr KLIMES, Filip PLESINGER, Josef HALAMEK, Vaclav KREMEN, Ivo VISCOR, Benjamin H. BRINKMANN, Martin PAIL, Milan BRÁZDIL, Gregory WORRELL and Pavel JURAK. Intracerebral EEG Artifact Identification Using Convolutional Neural Networks. NEUROINFORMATICS. TOTOWA: HUMANA PRESS INC, 2019, vol. 17, No 2, p. 225-234. ISSN 1539-2791. Available from: https://dx.doi.org/10.1007/s12021-018-9397-6.

@article{1531399,
   author = {Nejedly, Petr and Cimbalnik, Jan and Klimes, Petr and Plesinger, Filip and Halamek, Josef and Kremen, Vaclav and Viscor, Ivo and Brinkmann, Benjamin H. and Pail, Martin and Brázdil, Milan and Worrell, Gregory and Jurak, Pavel},
   article_location = {TOTOWA},
   article_number = {2},
   doi = {http://dx.doi.org/10.1007/s12021-018-9397-6},
   keywords = {Intracranial EEG (iEEG); Noise detection; Convolutional neural networks (CNN); Artifact probability matrix (APM)},
   language = {eng},
   issn = {1539-2791},
   journal = {NEUROINFORMATICS},
   title = {Intracerebral EEG Artifact Identification Using Convolutional Neural Networks},
   url = {https://link.springer.com/content/pdf/10.1007%2Fs12021-018-9397-6.pdf},
   volume = {17},
   year = {2019}
}

TY  - JOUR
ID  - 1531399
AU  - Nejedly, Petr - Cimbalnik, Jan - Klimes, Petr - Plesinger, Filip - Halamek, Josef - Kremen, Vaclav - Viscor, Ivo - Brinkmann, Benjamin H. - Pail, Martin - Brázdil, Milan - Worrell, Gregory - Jurak, Pavel
PY  - 2019
TI  - Intracerebral EEG Artifact Identification Using Convolutional Neural Networks
JF  - NEUROINFORMATICS
VL  - 17
IS  - 2
SP  - 225-234
EP  - 225-234
PB  - HUMANA PRESS INC
SN  - 15392791
KW  - Intracranial EEG (iEEG)
KW  - Noise detection
KW  - Convolutional neural networks (CNN)
KW  - Artifact probability matrix (APM)
UR  - https://link.springer.com/content/pdf/10.1007%2Fs12021-018-9397-6.pdf
L2  - https://link.springer.com/content/pdf/10.1007%2Fs12021-018-9397-6.pdf
N2  - Manual and semi-automatic identification of artifacts and unwanted physiological signals in large intracerebral electroencephalographic (iEEG) recordings is time consuming and inaccurate. To date, unsupervised methods to accurately detect iEEG artifacts are not available. This study introduces a novel machine-learning approach for detection of artifacts in iEEG signals in clinically controlled conditions using convolutional neural networks (CNN) and benchmarks the method's performance against expert annotations. The method was trained and tested on data obtained from St Anne's University Hospital (Brno, Czech Republic) and validated on data from Mayo Clinic (Rochester, Minnesota, U.S.A). We show that the proposed technique can be used as a generalized model for iEEG artifact detection. Moreover, a transfer learning process might be used for retraining of the generalized version to form a data-specific model. The generalized model can be efficiently retrained for use with different EEG acquisition systems and noise environments. The generalized and specialized model F1 scores on the testing dataset were 0.81 and 0.96, respectively. The CNN model provides faster, more objective, and more reproducible iEEG artifact detection compared to manual approaches.
ER  -

NEJEDLY, Petr, Jan CIMBALNIK, Petr KLIMES, Filip PLESINGER, Josef HALAMEK, Vaclav KREMEN, Ivo VISCOR, Benjamin H. BRINKMANN, Martin PAIL, Milan BRÁZDIL, Gregory WORRELL and Pavel JURAK. Intracerebral EEG Artifact Identification Using Convolutional Neural Networks. \textit{NEUROINFORMATICS}. TOTOWA: HUMANA PRESS INC, 2019, vol.~17, No~2, p.~225-234. ISSN~1539-2791. Available from: https://dx.doi.org/10.1007/s12021-018-9397-6.

Detailed Information on Publication Record