Computer-Aided Intracranial EEG Signal Identification Method Based on a Multi-Branch Deep Learning Fusion Model and Clinical Validation
Surgical intervention or the control of drug-refractory epilepsy requires accurate analysis of invasive inspection intracranial EEG (iEEG) data. A multi-branch deep learning fusion model is proposed to identify epileptogenic signals from the epileptogenic area of the brain. The classical approach ex...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-05-01
|
| Series: | Brain Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3425/11/5/615 |
| _version_ | 1797534557920034816 |
|---|---|
| author | Yiping Wang Yang Dai Zimo Liu Jinjie Guo Gongpeng Cao Mowei Ouyang Da Liu Yongzhi Shan Guixia Kang Guoguang Zhao |
| author_facet | Yiping Wang Yang Dai Zimo Liu Jinjie Guo Gongpeng Cao Mowei Ouyang Da Liu Yongzhi Shan Guixia Kang Guoguang Zhao |
| author_sort | Yiping Wang |
| collection | DOAJ |
| description | Surgical intervention or the control of drug-refractory epilepsy requires accurate analysis of invasive inspection intracranial EEG (iEEG) data. A multi-branch deep learning fusion model is proposed to identify epileptogenic signals from the epileptogenic area of the brain. The classical approach extracts multi-domain signal wave features to construct a time-series feature sequence and then abstracts it through the bi-directional long short-term memory attention machine (Bi-LSTM-AM) classifier. The deep learning approach uses raw time-series signals to build a one-dimensional convolutional neural network (1D-CNN) to achieve end-to-end deep feature extraction and signal detection. These two branches are integrated to obtain deep fusion features and results. Resampling is employed to split the imbalanced epileptogenic and non-epileptogenic samples into balanced subsets for clinical validation. The model is validated over two publicly available benchmark iEEG databases to verify its effectiveness on a private, large-scale, clinical stereo EEG database. The model achieves high sensitivity (97.78%), accuracy (97.60%), and specificity (97.42%) on the Bern–Barcelona database, surpassing the performance of existing state-of-the-art techniques. It is then demonstrated on a clinical dataset with an average intra-subject accuracy of 92.53% and cross-subject accuracy of 88.03%. The results suggest that the proposed method is a valuable and extremely robust approach to help researchers and clinicians develop an automated method to identify the source of iEEG signals. |
| first_indexed | 2024-03-10T11:32:11Z |
| format | Article |
| id | doaj.art-e47ce2fa31714911bf7d39cca1ea8ca9 |
| institution | Directory Open Access Journal |
| issn | 2076-3425 |
| language | English |
| last_indexed | 2024-03-10T11:32:11Z |
| publishDate | 2021-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Brain Sciences |
| spelling | doaj.art-e47ce2fa31714911bf7d39cca1ea8ca92023-11-21T19:12:11ZengMDPI AGBrain Sciences2076-34252021-05-0111561510.3390/brainsci11050615Computer-Aided Intracranial EEG Signal Identification Method Based on a Multi-Branch Deep Learning Fusion Model and Clinical ValidationYiping Wang0Yang Dai1Zimo Liu2Jinjie Guo3Gongpeng Cao4Mowei Ouyang5Da Liu6Yongzhi Shan7Guixia Kang8Guoguang Zhao9Key Laboratory of Universal Wireless Communications, Ministry of Education, Beijing University of Posts and Telecommunications, No. 10 Xitucheng Road, Haidian District, Beijing 100876, ChinaDepartment of Neurosurgery, Xuan Wu Hospital, Capital Medical University, No. 45 Changchun Street, Xicheng District, Beijing 100053, ChinaKey Laboratory of Universal Wireless Communications, Ministry of Education, Beijing University of Posts and Telecommunications, No. 10 Xitucheng Road, Haidian District, Beijing 100876, ChinaKey Laboratory of Universal Wireless Communications, Ministry of Education, Beijing University of Posts and Telecommunications, No. 10 Xitucheng Road, Haidian District, Beijing 100876, ChinaKey Laboratory of Universal Wireless Communications, Ministry of Education, Beijing University of Posts and Telecommunications, No. 10 Xitucheng Road, Haidian District, Beijing 100876, ChinaKey Laboratory of Universal Wireless Communications, Ministry of Education, Beijing University of Posts and Telecommunications, No. 10 Xitucheng Road, Haidian District, Beijing 100876, ChinaRobotics Institute, School of Mechanical Engineering & Automation, BeiHang University, No. 37 Xueyuan Road, Haidian District, Beijing 100191, ChinaDepartment of Neurosurgery, Xuan Wu Hospital, Capital Medical University, No. 45 Changchun Street, Xicheng District, Beijing 100053, ChinaKey Laboratory of Universal Wireless Communications, Ministry of Education, Beijing University of Posts and Telecommunications, No. 10 Xitucheng Road, Haidian District, Beijing 100876, ChinaDepartment of Neurosurgery, Xuan Wu Hospital, Capital Medical University, No. 45 Changchun Street, Xicheng District, Beijing 100053, ChinaSurgical intervention or the control of drug-refractory epilepsy requires accurate analysis of invasive inspection intracranial EEG (iEEG) data. A multi-branch deep learning fusion model is proposed to identify epileptogenic signals from the epileptogenic area of the brain. The classical approach extracts multi-domain signal wave features to construct a time-series feature sequence and then abstracts it through the bi-directional long short-term memory attention machine (Bi-LSTM-AM) classifier. The deep learning approach uses raw time-series signals to build a one-dimensional convolutional neural network (1D-CNN) to achieve end-to-end deep feature extraction and signal detection. These two branches are integrated to obtain deep fusion features and results. Resampling is employed to split the imbalanced epileptogenic and non-epileptogenic samples into balanced subsets for clinical validation. The model is validated over two publicly available benchmark iEEG databases to verify its effectiveness on a private, large-scale, clinical stereo EEG database. The model achieves high sensitivity (97.78%), accuracy (97.60%), and specificity (97.42%) on the Bern–Barcelona database, surpassing the performance of existing state-of-the-art techniques. It is then demonstrated on a clinical dataset with an average intra-subject accuracy of 92.53% and cross-subject accuracy of 88.03%. The results suggest that the proposed method is a valuable and extremely robust approach to help researchers and clinicians develop an automated method to identify the source of iEEG signals.https://www.mdpi.com/2076-3425/11/5/615intracranial EEG (iEEG)SEEGepileptogenic signals identificationmulti-branch deep learning fusion |
| spellingShingle | Yiping Wang Yang Dai Zimo Liu Jinjie Guo Gongpeng Cao Mowei Ouyang Da Liu Yongzhi Shan Guixia Kang Guoguang Zhao Computer-Aided Intracranial EEG Signal Identification Method Based on a Multi-Branch Deep Learning Fusion Model and Clinical Validation Brain Sciences intracranial EEG (iEEG) SEEG epileptogenic signals identification multi-branch deep learning fusion |
| title | Computer-Aided Intracranial EEG Signal Identification Method Based on a Multi-Branch Deep Learning Fusion Model and Clinical Validation |
| title_full | Computer-Aided Intracranial EEG Signal Identification Method Based on a Multi-Branch Deep Learning Fusion Model and Clinical Validation |
| title_fullStr | Computer-Aided Intracranial EEG Signal Identification Method Based on a Multi-Branch Deep Learning Fusion Model and Clinical Validation |
| title_full_unstemmed | Computer-Aided Intracranial EEG Signal Identification Method Based on a Multi-Branch Deep Learning Fusion Model and Clinical Validation |
| title_short | Computer-Aided Intracranial EEG Signal Identification Method Based on a Multi-Branch Deep Learning Fusion Model and Clinical Validation |
| title_sort | computer aided intracranial eeg signal identification method based on a multi branch deep learning fusion model and clinical validation |
| topic | intracranial EEG (iEEG) SEEG epileptogenic signals identification multi-branch deep learning fusion |
| url | https://www.mdpi.com/2076-3425/11/5/615 |
| work_keys_str_mv | AT yipingwang computeraidedintracranialeegsignalidentificationmethodbasedonamultibranchdeeplearningfusionmodelandclinicalvalidation AT yangdai computeraidedintracranialeegsignalidentificationmethodbasedonamultibranchdeeplearningfusionmodelandclinicalvalidation AT zimoliu computeraidedintracranialeegsignalidentificationmethodbasedonamultibranchdeeplearningfusionmodelandclinicalvalidation AT jinjieguo computeraidedintracranialeegsignalidentificationmethodbasedonamultibranchdeeplearningfusionmodelandclinicalvalidation AT gongpengcao computeraidedintracranialeegsignalidentificationmethodbasedonamultibranchdeeplearningfusionmodelandclinicalvalidation AT moweiouyang computeraidedintracranialeegsignalidentificationmethodbasedonamultibranchdeeplearningfusionmodelandclinicalvalidation AT daliu computeraidedintracranialeegsignalidentificationmethodbasedonamultibranchdeeplearningfusionmodelandclinicalvalidation AT yongzhishan computeraidedintracranialeegsignalidentificationmethodbasedonamultibranchdeeplearningfusionmodelandclinicalvalidation AT guixiakang computeraidedintracranialeegsignalidentificationmethodbasedonamultibranchdeeplearningfusionmodelandclinicalvalidation AT guoguangzhao computeraidedintracranialeegsignalidentificationmethodbasedonamultibranchdeeplearningfusionmodelandclinicalvalidation |