A lightweight convolutional neural network for assessing an EEG risk marker for sudden unexpected death in epilepsy
Abstract Background Convolutional neural network (CNN) has achieved state-of-art performance in many electroencephalogram (EEG) related studies. However, the application of CNN in prediction of risk factors for sudden unexpected death in epilepsy (SUDEP) remains as an underexplored area. It is uncle...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2020-12-01
|
| Series: | BMC Medical Informatics and Decision Making |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12911-020-01310-y |
| _version_ | 1819295547009269760 |
|---|---|
| author | Cong Zhu Yejin Kim Xiaoqian Jiang Samden Lhatoo Hampson Jaison Guo-Qiang Zhang |
| author_facet | Cong Zhu Yejin Kim Xiaoqian Jiang Samden Lhatoo Hampson Jaison Guo-Qiang Zhang |
| author_sort | Cong Zhu |
| collection | DOAJ |
| description | Abstract Background Convolutional neural network (CNN) has achieved state-of-art performance in many electroencephalogram (EEG) related studies. However, the application of CNN in prediction of risk factors for sudden unexpected death in epilepsy (SUDEP) remains as an underexplored area. It is unclear how the trade-off between computation cost and prediction power varies with changes in the complexity and depth of neural nets. Methods The purpose of this study was to explore the feasibility of using a lightweight CNN to predict SUDEP. A total of 170 patients were included in the analyses. The CNN model was trained using clips with 10-s signals sampled from the original EEG. We implemented Hann function to smooth the raw EEG signal and evaluated its effect by choosing different strength of denoising filter. In addition, we experimented two variations of the proposed model: (1) converting EEG input into an “RGB” format to address EEG channels underlying spatial correlation and (2) incorporating residual network (ResNet) into the bottle neck position of the proposed structure of baseline CNN. Results The proposed baseline CNN model with lightweight architecture achieved the best AUC of 0.72. A moderate noise removal step facilitated the training of CNN model by ensuring stability of performance. We did not observe further improvement in model’s accuracy by increasing the strength of denoising filter. Conclusion Post-seizure slow activity in EEG is a potential marker for SUDEP, our proposed lightweight architecture of CNN achieved satisfying trade-off between efficiently identifying such biomarker and computational cost. It also has a flexible interface to be integrated with different variations in structure leaving room for further improvement of the model’s performance in automating EEG signal annotation. |
| first_indexed | 2024-12-24T04:43:57Z |
| format | Article |
| id | doaj.art-84d322a0bbca4ed090163da50fbe6be8 |
| institution | Directory Open Access Journal |
| issn | 1472-6947 |
| language | English |
| last_indexed | 2024-12-24T04:43:57Z |
| publishDate | 2020-12-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Informatics and Decision Making |
| spelling | doaj.art-84d322a0bbca4ed090163da50fbe6be82022-12-21T17:14:44ZengBMCBMC Medical Informatics and Decision Making1472-69472020-12-0120S121810.1186/s12911-020-01310-yA lightweight convolutional neural network for assessing an EEG risk marker for sudden unexpected death in epilepsyCong Zhu0Yejin Kim1Xiaoqian Jiang2Samden Lhatoo3Hampson Jaison4Guo-Qiang Zhang5Department of Epidemiology, Human Genetics and Environmental Sciences, School of Public Health, University of Texas Health Science Center at HoustonSchool of Biomedical Informatics, University of Texas Health Science Center at HoustonSchool of Biomedical Informatics, University of Texas Health Science Center at HoustonDepartment of Neurology, McGovern Medical School, University of Texas Health Science Center at HoustonDepartment of Neurology, McGovern Medical School, University of Texas Health Science Center at HoustonDepartment of Neurology, McGovern Medical School, University of Texas Health Science Center at HoustonAbstract Background Convolutional neural network (CNN) has achieved state-of-art performance in many electroencephalogram (EEG) related studies. However, the application of CNN in prediction of risk factors for sudden unexpected death in epilepsy (SUDEP) remains as an underexplored area. It is unclear how the trade-off between computation cost and prediction power varies with changes in the complexity and depth of neural nets. Methods The purpose of this study was to explore the feasibility of using a lightweight CNN to predict SUDEP. A total of 170 patients were included in the analyses. The CNN model was trained using clips with 10-s signals sampled from the original EEG. We implemented Hann function to smooth the raw EEG signal and evaluated its effect by choosing different strength of denoising filter. In addition, we experimented two variations of the proposed model: (1) converting EEG input into an “RGB” format to address EEG channels underlying spatial correlation and (2) incorporating residual network (ResNet) into the bottle neck position of the proposed structure of baseline CNN. Results The proposed baseline CNN model with lightweight architecture achieved the best AUC of 0.72. A moderate noise removal step facilitated the training of CNN model by ensuring stability of performance. We did not observe further improvement in model’s accuracy by increasing the strength of denoising filter. Conclusion Post-seizure slow activity in EEG is a potential marker for SUDEP, our proposed lightweight architecture of CNN achieved satisfying trade-off between efficiently identifying such biomarker and computational cost. It also has a flexible interface to be integrated with different variations in structure leaving room for further improvement of the model’s performance in automating EEG signal annotation.https://doi.org/10.1186/s12911-020-01310-yConvolutional neural networkSudden death in epilepsyPGESEEG suppressionDeep learning |
| spellingShingle | Cong Zhu Yejin Kim Xiaoqian Jiang Samden Lhatoo Hampson Jaison Guo-Qiang Zhang A lightweight convolutional neural network for assessing an EEG risk marker for sudden unexpected death in epilepsy BMC Medical Informatics and Decision Making Convolutional neural network Sudden death in epilepsy PGES EEG suppression Deep learning |
| title | A lightweight convolutional neural network for assessing an EEG risk marker for sudden unexpected death in epilepsy |
| title_full | A lightweight convolutional neural network for assessing an EEG risk marker for sudden unexpected death in epilepsy |
| title_fullStr | A lightweight convolutional neural network for assessing an EEG risk marker for sudden unexpected death in epilepsy |
| title_full_unstemmed | A lightweight convolutional neural network for assessing an EEG risk marker for sudden unexpected death in epilepsy |
| title_short | A lightweight convolutional neural network for assessing an EEG risk marker for sudden unexpected death in epilepsy |
| title_sort | lightweight convolutional neural network for assessing an eeg risk marker for sudden unexpected death in epilepsy |
| topic | Convolutional neural network Sudden death in epilepsy PGES EEG suppression Deep learning |
| url | https://doi.org/10.1186/s12911-020-01310-y |
| work_keys_str_mv | AT congzhu alightweightconvolutionalneuralnetworkforassessinganeegriskmarkerforsuddenunexpecteddeathinepilepsy AT yejinkim alightweightconvolutionalneuralnetworkforassessinganeegriskmarkerforsuddenunexpecteddeathinepilepsy AT xiaoqianjiang alightweightconvolutionalneuralnetworkforassessinganeegriskmarkerforsuddenunexpecteddeathinepilepsy AT samdenlhatoo alightweightconvolutionalneuralnetworkforassessinganeegriskmarkerforsuddenunexpecteddeathinepilepsy AT hampsonjaison alightweightconvolutionalneuralnetworkforassessinganeegriskmarkerforsuddenunexpecteddeathinepilepsy AT guoqiangzhang alightweightconvolutionalneuralnetworkforassessinganeegriskmarkerforsuddenunexpecteddeathinepilepsy AT congzhu lightweightconvolutionalneuralnetworkforassessinganeegriskmarkerforsuddenunexpecteddeathinepilepsy AT yejinkim lightweightconvolutionalneuralnetworkforassessinganeegriskmarkerforsuddenunexpecteddeathinepilepsy AT xiaoqianjiang lightweightconvolutionalneuralnetworkforassessinganeegriskmarkerforsuddenunexpecteddeathinepilepsy AT samdenlhatoo lightweightconvolutionalneuralnetworkforassessinganeegriskmarkerforsuddenunexpecteddeathinepilepsy AT hampsonjaison lightweightconvolutionalneuralnetworkforassessinganeegriskmarkerforsuddenunexpecteddeathinepilepsy AT guoqiangzhang lightweightconvolutionalneuralnetworkforassessinganeegriskmarkerforsuddenunexpecteddeathinepilepsy |