Automatic Seizure Classification Based on Domain-Invariant Deep Representation of EEG
Accurate identification of the type of seizure is very important for the treatment plan and drug prescription of epileptic patients. Artificial intelligence has shown considerable potential in the fields of automated EEG analysis and seizure classification. However, the highly personalized represent...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2021-10-01
|
| Series: | Frontiers in Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnins.2021.760987/full |
| _version_ | 1819022391932616704 |
|---|---|
| author | Xincheng Cao Xincheng Cao Bin Yao Bin Yao Binqiang Chen Binqiang Chen Weifang Sun Guowei Tan |
| author_facet | Xincheng Cao Xincheng Cao Bin Yao Bin Yao Binqiang Chen Binqiang Chen Weifang Sun Guowei Tan |
| author_sort | Xincheng Cao |
| collection | DOAJ |
| description | Accurate identification of the type of seizure is very important for the treatment plan and drug prescription of epileptic patients. Artificial intelligence has shown considerable potential in the fields of automated EEG analysis and seizure classification. However, the highly personalized representation of epileptic seizures in EEG has led to many research results that are not satisfactory in clinical applications. In order to improve the clinical adaptability of the algorithm, this paper proposes an adversarial learning-driven domain-invariant deep feature representation method, which enables the hybrid deep networks (HDN) to reliably identify seizure types. In the train phase, we first use the labeled multi-lead EEG short samples to train squeeze-and-excitation networks (SENet) to extract short-term features, and then use the compressed samples to train the long short-term memory networks (LSTM) to extract long-time features and construct a classifier. In the inference phase, we first adjust the feature mapping of LSTM through the adversarial learning between LSTM and clustering subnet so that the EEG of the target patient and the EEG in the database obey the same distribution in the deep feature space. Finally, we use the adjusted classifier to identify the type of seizure. Experiments were carried out based on the TUH EEG Seizure Corpus and CHB-MIT seizure database. The experimental results show that the proposed domain adaptive deep feature representation improves the classification accuracy of the hybrid deep model in the target set by 5%. It is of great significance for the clinical application of EEG automatic analysis equipment. |
| first_indexed | 2024-12-21T04:22:16Z |
| format | Article |
| id | doaj.art-8e0b1c89b1f141088b5e3f062fd2f3c6 |
| institution | Directory Open Access Journal |
| issn | 1662-453X |
| language | English |
| last_indexed | 2024-12-21T04:22:16Z |
| publishDate | 2021-10-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroscience |
| spelling | doaj.art-8e0b1c89b1f141088b5e3f062fd2f3c62022-12-21T19:16:09ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2021-10-011510.3389/fnins.2021.760987760987Automatic Seizure Classification Based on Domain-Invariant Deep Representation of EEGXincheng Cao0Xincheng Cao1Bin Yao2Bin Yao3Binqiang Chen4Binqiang Chen5Weifang Sun6Guowei Tan7School of Aerospace Engineering, Xiamen University, Xiamen, ChinaShenzhen Research Institute of Xiamen University, Shenzhen, ChinaSchool of Aerospace Engineering, Xiamen University, Xiamen, ChinaShenzhen Research Institute of Xiamen University, Shenzhen, ChinaSchool of Aerospace Engineering, Xiamen University, Xiamen, ChinaShenzhen Research Institute of Xiamen University, Shenzhen, ChinaCollege of Mechanical and Electrical Engineering, Wenzhou University, Wenzhou, ChinaXiamen Key Laboratory of Brain Center, Department of Neurosurgery, The First Affiliated Hospital of Xiamen University, Xiamen, ChinaAccurate identification of the type of seizure is very important for the treatment plan and drug prescription of epileptic patients. Artificial intelligence has shown considerable potential in the fields of automated EEG analysis and seizure classification. However, the highly personalized representation of epileptic seizures in EEG has led to many research results that are not satisfactory in clinical applications. In order to improve the clinical adaptability of the algorithm, this paper proposes an adversarial learning-driven domain-invariant deep feature representation method, which enables the hybrid deep networks (HDN) to reliably identify seizure types. In the train phase, we first use the labeled multi-lead EEG short samples to train squeeze-and-excitation networks (SENet) to extract short-term features, and then use the compressed samples to train the long short-term memory networks (LSTM) to extract long-time features and construct a classifier. In the inference phase, we first adjust the feature mapping of LSTM through the adversarial learning between LSTM and clustering subnet so that the EEG of the target patient and the EEG in the database obey the same distribution in the deep feature space. Finally, we use the adjusted classifier to identify the type of seizure. Experiments were carried out based on the TUH EEG Seizure Corpus and CHB-MIT seizure database. The experimental results show that the proposed domain adaptive deep feature representation improves the classification accuracy of the hybrid deep model in the target set by 5%. It is of great significance for the clinical application of EEG automatic analysis equipment.https://www.frontiersin.org/articles/10.3389/fnins.2021.760987/fullelectroencephalographyseizure classificationdeep learningdomain-invariant representationhybrid deep model |
| spellingShingle | Xincheng Cao Xincheng Cao Bin Yao Bin Yao Binqiang Chen Binqiang Chen Weifang Sun Guowei Tan Automatic Seizure Classification Based on Domain-Invariant Deep Representation of EEG Frontiers in Neuroscience electroencephalography seizure classification deep learning domain-invariant representation hybrid deep model |
| title | Automatic Seizure Classification Based on Domain-Invariant Deep Representation of EEG |
| title_full | Automatic Seizure Classification Based on Domain-Invariant Deep Representation of EEG |
| title_fullStr | Automatic Seizure Classification Based on Domain-Invariant Deep Representation of EEG |
| title_full_unstemmed | Automatic Seizure Classification Based on Domain-Invariant Deep Representation of EEG |
| title_short | Automatic Seizure Classification Based on Domain-Invariant Deep Representation of EEG |
| title_sort | automatic seizure classification based on domain invariant deep representation of eeg |
| topic | electroencephalography seizure classification deep learning domain-invariant representation hybrid deep model |
| url | https://www.frontiersin.org/articles/10.3389/fnins.2021.760987/full |
| work_keys_str_mv | AT xinchengcao automaticseizureclassificationbasedondomaininvariantdeeprepresentationofeeg AT xinchengcao automaticseizureclassificationbasedondomaininvariantdeeprepresentationofeeg AT binyao automaticseizureclassificationbasedondomaininvariantdeeprepresentationofeeg AT binyao automaticseizureclassificationbasedondomaininvariantdeeprepresentationofeeg AT binqiangchen automaticseizureclassificationbasedondomaininvariantdeeprepresentationofeeg AT binqiangchen automaticseizureclassificationbasedondomaininvariantdeeprepresentationofeeg AT weifangsun automaticseizureclassificationbasedondomaininvariantdeeprepresentationofeeg AT guoweitan automaticseizureclassificationbasedondomaininvariantdeeprepresentationofeeg |