<abstract xmlns="http://www.w3.org/1999/xhtml"><p>Deep brain stimulation (DBS) is an internationally accepted form of treatment option for selected patients with Parkinson’s disease and dystonia. Intraoperative extracellular microelectrode recordings (MER) are considered as the standard electrophysiological method for the precise positioning of the DBS electrode into the target brain structure. Pre-processing of MERs is a key phase in clinical analysis, with intraoperative microelectrode recordings being prone to several artifact groups (up to 25 %). The aim of this methodological article is to provide a convolutional neural network (CNN) processing pipeline for the detection of artifacts in an MER. We applied continuous wavelet transform (CWT) to generate an over-complete time–frequency representation. We demonstrated that when attempting to find artifacts in an MER, the new CNN + CWT provides a high level of accuracy (ACC = 88.1 %), identifies individual classes of artifacts (ACC = 75.3 %) and also offers artifact time onset detail, which can lead to a reduction in false positives/negatives. In summary, the presented methodology is capable of identifying and removing various artifacts in a comprehensive database of MER and represents a substantial improvement over the existing methodology. We believe that this approach will assist in the proposal of interesting clinical hypotheses and will have neurologically relevant effects.</p></abstract>

Deep brain stimulation (DBS) is an internationally accepted form of treatment option for selected patients with Parkinson’s disease and dystonia. Intraoperative extracellular microelectrode recordings (MER) are considered as the standard electrophysiological method for the precise positioning of the DBS electrode into the target brain structure. Pre-processing of MERs is a key phase in clinical analysis, with intraoperative microelectrode recordings being prone to several artifact groups (up to 25 %). The aim of this methodological article is to provide a convolutional neural network (CNN) processing pipeline for the detection of artifacts in an MER. We applied continuous wavelet transform (CWT) to generate an over-complete time–frequency representation. We demonstrated that when attempting to find artifacts in an MER, the new CNN + CWT provides a high level of accuracy (ACC = 88.1 %), identifies individual classes of artifacts (ACC = 75.3 %) and also offers artifact time onset detail, which can lead to a reduction in false positives/negatives. In summary, the presented methodology is capable of identifying and removing various artifacts in a comprehensive database of MER and represents a substantial improvement over the existing methodology. We believe that this approach will assist in the proposal of interesting clinical hypotheses and will have neurologically relevant effects.

Identification of Microrecording Artifacts with Wavelet Analysis and Convolutional Neural Network: An Image Recognition Approach

Department of Biomedical Informatics, Faculty of Biomedical Engineering

Department of Cybernetics, Faculty of Electrical Engineering

Department of Neurology and Center of Clinical Neuroscience

First Faculty of Medicine and General University Hospital, Charles University

Measurement Science Review

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Deep brain stimulation (DBS) is an internationally accepted form of treatment option for selected patients with Parkinson’s disease and dystonia....

Identification of Microrecording Artifacts with Wavelet Analysis and Convolutional Neural Network: An Image Recognition Approach

Published Online: Oct 07, 2019

Page range: 222 - 231

Received: Mar 20, 2019

Accepted: Aug 30, 2019

DOI: https://doi.org/10.2478/msr-2019-0029

Keywords
Artifacts detection, convolutional neural networks, Parkinson’s disease, deep brain stimulation, microrecording, wavelet analysis

© 2019 Ondřej Klempíř et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Identification of Microrecording Artifacts with Wavelet Analysis and Convolutional Neural Network: An Image Recognition Approach

Published Online: Oct 07, 2019

Page range: 222 - 231

Received: Mar 20, 2019

Accepted: Aug 30, 2019

DOI: https://doi.org/10.2478/msr-2019-0029

KeywordsArtifacts detection, convolutional neural networks, Parkinson’s disease, deep brain stimulation, microrecording, wavelet analysis

© 2019 Ondřej Klempíř et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Keywords
Artifacts detection, convolutional neural networks, Parkinson’s disease, deep brain stimulation, microrecording, wavelet analysis