Automatic Seizure Detection Based on Stockwell Transform and Transformer
Epilepsy is a chronic neurological disease associated with abnormal neuronal activity in the brain. Seizure detection algorithms are essential in reducing the workload of medical staff reviewing electroencephalogram (EEG) records. In this work, we propose a novel automatic epileptic EEG detection me...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/24/1/77 |
| _version_ | 1797358212943446016 |
|---|---|
| author | Xiangwen Zhong Guoyang Liu Xingchen Dong Chuanyu Li Haotian Li Haozhou Cui Weidong Zhou |
| author_facet | Xiangwen Zhong Guoyang Liu Xingchen Dong Chuanyu Li Haotian Li Haozhou Cui Weidong Zhou |
| author_sort | Xiangwen Zhong |
| collection | DOAJ |
| description | Epilepsy is a chronic neurological disease associated with abnormal neuronal activity in the brain. Seizure detection algorithms are essential in reducing the workload of medical staff reviewing electroencephalogram (EEG) records. In this work, we propose a novel automatic epileptic EEG detection method based on Stockwell transform and Transformer. First, the S-transform is applied to the original EEG segments, acquiring accurate time-frequency representations. Subsequently, the obtained time-frequency matrices are grouped into different EEG rhythm blocks and compressed as vectors in these EEG sub-bands. After that, these feature vectors are fed into the Transformer network for feature selection and classification. Moreover, a series of post-processing methods were introduced to enhance the efficiency of the system. When evaluating the public CHB-MIT database, the proposed algorithm achieved an accuracy of 96.15%, a sensitivity of 96.11%, a specificity of 96.38%, a precision of 96.33%, and an area under the curve (AUC) of 0.98 in segment-based experiments, along with a sensitivity of 96.57%, a false detection rate of 0.38/h, and a delay of 20.62 s in event-based experiments. These outstanding results demonstrate the feasibility of implementing this seizure detection method in future clinical applications. |
| first_indexed | 2024-03-08T14:57:38Z |
| format | Article |
| id | doaj.art-ac69e8ff1a7c4b708171afe7356ba375 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-08T14:57:38Z |
| publishDate | 2023-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-ac69e8ff1a7c4b708171afe7356ba3752024-01-10T15:08:28ZengMDPI AGSensors1424-82202023-12-012417710.3390/s24010077Automatic Seizure Detection Based on Stockwell Transform and TransformerXiangwen Zhong0Guoyang Liu1Xingchen Dong2Chuanyu Li3Haotian Li4Haozhou Cui5Weidong Zhou6School of Integrated Circuits, Shandong University, Jinan 260100, ChinaSchool of Integrated Circuits, Shandong University, Jinan 260100, ChinaSchool of Integrated Circuits, Shandong University, Jinan 260100, ChinaSchool of Integrated Circuits, Shandong University, Jinan 260100, ChinaSchool of Integrated Circuits, Shandong University, Jinan 260100, ChinaSchool of Integrated Circuits, Shandong University, Jinan 260100, ChinaSchool of Integrated Circuits, Shandong University, Jinan 260100, ChinaEpilepsy is a chronic neurological disease associated with abnormal neuronal activity in the brain. Seizure detection algorithms are essential in reducing the workload of medical staff reviewing electroencephalogram (EEG) records. In this work, we propose a novel automatic epileptic EEG detection method based on Stockwell transform and Transformer. First, the S-transform is applied to the original EEG segments, acquiring accurate time-frequency representations. Subsequently, the obtained time-frequency matrices are grouped into different EEG rhythm blocks and compressed as vectors in these EEG sub-bands. After that, these feature vectors are fed into the Transformer network for feature selection and classification. Moreover, a series of post-processing methods were introduced to enhance the efficiency of the system. When evaluating the public CHB-MIT database, the proposed algorithm achieved an accuracy of 96.15%, a sensitivity of 96.11%, a specificity of 96.38%, a precision of 96.33%, and an area under the curve (AUC) of 0.98 in segment-based experiments, along with a sensitivity of 96.57%, a false detection rate of 0.38/h, and a delay of 20.62 s in event-based experiments. These outstanding results demonstrate the feasibility of implementing this seizure detection method in future clinical applications.https://www.mdpi.com/1424-8220/24/1/77automatic seizure detectiontransformerstockwell transformEEG |
| spellingShingle | Xiangwen Zhong Guoyang Liu Xingchen Dong Chuanyu Li Haotian Li Haozhou Cui Weidong Zhou Automatic Seizure Detection Based on Stockwell Transform and Transformer Sensors automatic seizure detection transformer stockwell transform EEG |
| title | Automatic Seizure Detection Based on Stockwell Transform and Transformer |
| title_full | Automatic Seizure Detection Based on Stockwell Transform and Transformer |
| title_fullStr | Automatic Seizure Detection Based on Stockwell Transform and Transformer |
| title_full_unstemmed | Automatic Seizure Detection Based on Stockwell Transform and Transformer |
| title_short | Automatic Seizure Detection Based on Stockwell Transform and Transformer |
| title_sort | automatic seizure detection based on stockwell transform and transformer |
| topic | automatic seizure detection transformer stockwell transform EEG |
| url | https://www.mdpi.com/1424-8220/24/1/77 |
| work_keys_str_mv | AT xiangwenzhong automaticseizuredetectionbasedonstockwelltransformandtransformer AT guoyangliu automaticseizuredetectionbasedonstockwelltransformandtransformer AT xingchendong automaticseizuredetectionbasedonstockwelltransformandtransformer AT chuanyuli automaticseizuredetectionbasedonstockwelltransformandtransformer AT haotianli automaticseizuredetectionbasedonstockwelltransformandtransformer AT haozhoucui automaticseizuredetectionbasedonstockwelltransformandtransformer AT weidongzhou automaticseizuredetectionbasedonstockwelltransformandtransformer |