TY - GEN
T1 - EEG Channel Interpolation Using Deep Encoder-decoder Networks
AU - Saba-Sadiya, Sari
AU - Alhanai, Tuka
AU - Liu, Taosheng
AU - Ghassemi, Mohammad M.
N1 - Funding Information:
We would like to thank the authors of EEG During Mental Arithmetic Data Set [20] for making their data available. This work was partially supported by the Michigan State University Graduate Office Fellowship.
Publisher Copyright:
© 2020 IEEE.
PY - 2020/12/16
Y1 - 2020/12/16
N2 - Electrode 'pop' artifacts originate from the spontaneous loss of connectivity between a surface and an electrode. Electroencephalography (EEG) uses a dense array of electrodes, hence popped segments are among the most pervasive type of artifact seen during the collection of EEG data. In many cases, the continuity of EEG data is critical for downstream applications (e.g. brain machine interface) and requires that popped segments be accurately interpolated. In this paper we frame the interpolation problem as a self-learning task using a deep encoder-decoder network. We compare our approach against contemporary interpolation methods on a publicly available EEG data set. Our approach exhibited a minimum of 15% improvement over contemporary approaches when tested on subjects and tasks not used during model training. We demonstrate how our model's performance can be enhanced further on novel subjects and tasks using transfer learning. All code and data associated with this study is open-source to enable ease of extension and practical use. To our knowledge, this work is the first solution to the EEG interpolation problem that uses deep learning.
AB - Electrode 'pop' artifacts originate from the spontaneous loss of connectivity between a surface and an electrode. Electroencephalography (EEG) uses a dense array of electrodes, hence popped segments are among the most pervasive type of artifact seen during the collection of EEG data. In many cases, the continuity of EEG data is critical for downstream applications (e.g. brain machine interface) and requires that popped segments be accurately interpolated. In this paper we frame the interpolation problem as a self-learning task using a deep encoder-decoder network. We compare our approach against contemporary interpolation methods on a publicly available EEG data set. Our approach exhibited a minimum of 15% improvement over contemporary approaches when tested on subjects and tasks not used during model training. We demonstrate how our model's performance can be enhanced further on novel subjects and tasks using transfer learning. All code and data associated with this study is open-source to enable ease of extension and practical use. To our knowledge, this work is the first solution to the EEG interpolation problem that uses deep learning.
UR - http://www.scopus.com/inward/record.url?scp=85100354289&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85100354289&partnerID=8YFLogxK
U2 - 10.1109/BIBM49941.2020.9312979
DO - 10.1109/BIBM49941.2020.9312979
M3 - Conference contribution
AN - SCOPUS:85100354289
T3 - Proceedings - 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020
SP - 2432
EP - 2439
BT - Proceedings - 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020
A2 - Park, Taesung
A2 - Cho, Young-Rae
A2 - Hu, Xiaohua Tony
A2 - Yoo, Illhoi
A2 - Woo, Hyun Goo
A2 - Wang, Jianxin
A2 - Facelli, Julio
A2 - Nam, Seungyoon
A2 - Kang, Mingon
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020
Y2 - 16 December 2020 through 19 December 2020
ER -