TY - GEN
T1 - Deep Learning-Based Automated Classification of Epileptic and Non-epileptic Scalp-EEG Signals
AU - Prabhu, Pooja
AU - Kotegar, Karunakar A.
AU - Mariyappa, N.
AU - Anitha, H.
AU - Bhargava, G. K.
AU - Saini, Jitender
AU - Sinha, Sanjib
N1 - Funding Information:
Acknowledgements This project was supported by CSIR-SRF under the file number 08/0602(11528)/2020-EMR-I. We thank the Manipal Institute of Technology for providing the computational facility.
Funding Information:
This project was supported by CSIR-SRF under the file number 08/0602(11528)/2020-EMR-I. We thank the Manipal Institute of Technology for providing the computational facility.
Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
PY - 2022
Y1 - 2022
N2 - Epilepsy is the most common neurological disorder that has affected 50 million worldwide population. Epilepsy is a chronic condition that is characterized as recurrent and unprovoked seizures. A seizure is defined as abnormal, excessive paroxysmal discharge of the cerebral neurons. In general, epilepsy evaluation, the high time resolution (~1–2 ms) electrophysiological data recorded using electroencephalography (EEG) is commonly used. The EEG data are visually inspected for epileptic seizure instances. However, the manual interpretation may lead to subjective error-causing misdiagnosis, and identifying the seizure instances in high temporal resolution data may be time-consuming. To mitigate these problems, the proposed study uses a hand-crafted deep learning model to classify the epileptic and non-epileptic EEG data. This study used EEG data acquired from two databases—University of Bonn and Physiobank Children’s Hospital Boston—Massachusetts Institute of Technology (CHB-MIT). We split the EEG data in 80:20 ratio for training and testing the model. The model was trained under three network specifications, which were designed based on the number of high-level features and low-level features. The result showed that the proposed study successfully classified epileptic and non-epileptic signals with the accuracy of 67% and 92% for University of Bonn data and CHB-MIT EEG data, respectively, for the network specification which had a high number of low-level features. We tested the model for optimal value of number of epochs (20) and learning rate (0.001). The study shows that the MIT-CHB data are suitable for classification as they have a good number of samples and balanced epileptic and non-epileptic signals.
AB - Epilepsy is the most common neurological disorder that has affected 50 million worldwide population. Epilepsy is a chronic condition that is characterized as recurrent and unprovoked seizures. A seizure is defined as abnormal, excessive paroxysmal discharge of the cerebral neurons. In general, epilepsy evaluation, the high time resolution (~1–2 ms) electrophysiological data recorded using electroencephalography (EEG) is commonly used. The EEG data are visually inspected for epileptic seizure instances. However, the manual interpretation may lead to subjective error-causing misdiagnosis, and identifying the seizure instances in high temporal resolution data may be time-consuming. To mitigate these problems, the proposed study uses a hand-crafted deep learning model to classify the epileptic and non-epileptic EEG data. This study used EEG data acquired from two databases—University of Bonn and Physiobank Children’s Hospital Boston—Massachusetts Institute of Technology (CHB-MIT). We split the EEG data in 80:20 ratio for training and testing the model. The model was trained under three network specifications, which were designed based on the number of high-level features and low-level features. The result showed that the proposed study successfully classified epileptic and non-epileptic signals with the accuracy of 67% and 92% for University of Bonn data and CHB-MIT EEG data, respectively, for the network specification which had a high number of low-level features. We tested the model for optimal value of number of epochs (20) and learning rate (0.001). The study shows that the MIT-CHB data are suitable for classification as they have a good number of samples and balanced epileptic and non-epileptic signals.
UR - http://www.scopus.com/inward/record.url?scp=85125243227&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85125243227&partnerID=8YFLogxK
U2 - 10.1007/978-981-16-7996-4_30
DO - 10.1007/978-981-16-7996-4_30
M3 - Conference contribution
AN - SCOPUS:85125243227
SN - 9789811679957
T3 - Smart Innovation, Systems and Technologies
SP - 425
EP - 435
BT - Machine Learning and Autonomous Systems - Proceedings of ICMLAS 2021
A2 - Chen, Joy Iong-Zong
A2 - Wang, Haoxiang
A2 - Du, Ke-Lin
A2 - Suma, V.
PB - Springer Science and Business Media Deutschland GmbH
T2 - International Conference on Machine Learning and Autonomous Systems, ICMLAS 2021
Y2 - 24 September 2021 through 25 September 2021
ER -