Unsupervised Feature Representation Based on Deep Boltzmann Machine for Seizure Detection
The Electroencephalogram (EEG) pattern of seizure activities is highly individual-dependent and requires experienced specialists to annotate seizure events. It is clinically time-consuming and error-prone to identify seizure activities by visually scanning EEG signals. Since EEG data are heavily und...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10064189/ |
| _version_ | 1797805103816638464 |
|---|---|
| author | Tengzi Liu Muhammad Zohaib Hassan Shah Xucun Yan Dongping Yang |
| author_facet | Tengzi Liu Muhammad Zohaib Hassan Shah Xucun Yan Dongping Yang |
| author_sort | Tengzi Liu |
| collection | DOAJ |
| description | The Electroencephalogram (EEG) pattern of seizure activities is highly individual-dependent and requires experienced specialists to annotate seizure events. It is clinically time-consuming and error-prone to identify seizure activities by visually scanning EEG signals. Since EEG data are heavily under-represented, supervised learning techniques are not always practical, particularly when the data is not sufficiently labelled. Visualization of EEG data in low-dimensional feature space can ease the annotation to support subsequent supervised learning for seizure detection. Here, we leverage the benefit of both the time-frequency domain features and the Deep Boltzmann Machine (DBM) based unsupervised learning techniques to represent EEG signals in a 2-dimensional (2D) feature space. A novel unsupervised learning approach based on DBM, namely DBM_transient, is proposed by training DBM to a transient state for representing EEG signals in a 2D feature space and clustering seizure and non-seizure events visually. The effectiveness of DBM_transient is demonstrated on a widely-used benchmark dataset from Bonn University (Bonn dataset) and a raw clinical dataset from Chinese 301 Hospital (C301 dataset), with a large fisher discriminant value, surpassing the abilities of other dimensionality reduction methods, including DBM converged to an equilibrium state, Kernel Principal Component Analysis, Isometric Feature Mapping, t-distributed Stochastic Neighbour Embedding, Uniform Manifold Approximation. Such feature representation and visualization can help physicians to understand better the normal versus epileptic brain activities of each patient and thus enhance their diagnosis and treatment abilities. The significance of our approach facilitates its future usage in clinical applications. |
| first_indexed | 2024-03-13T05:46:33Z |
| format | Article |
| id | doaj.art-a38d7623a04f4ee78ddb111a9f06f2b3 |
| institution | Directory Open Access Journal |
| issn | 1558-0210 |
| language | English |
| last_indexed | 2024-03-13T05:46:33Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
| spelling | doaj.art-a38d7623a04f4ee78ddb111a9f06f2b32023-06-13T20:09:24ZengIEEEIEEE Transactions on Neural Systems and Rehabilitation Engineering1558-02102023-01-01311624163410.1109/TNSRE.2023.325382110064189Unsupervised Feature Representation Based on Deep Boltzmann Machine for Seizure DetectionTengzi Liu0https://orcid.org/0000-0001-9174-9049Muhammad Zohaib Hassan Shah1https://orcid.org/0000-0002-8817-5766Xucun Yan2Dongping Yang3https://orcid.org/0000-0002-1597-5201Research Center for Augmented Intelligence, Research Institute of Artificial Intelligence, Zhejiang Laboratory, Hangzhou, ChinaResearch Center for Augmented Intelligence, Research Institute of Artificial Intelligence, Zhejiang Laboratory, Hangzhou, ChinaSchool of Electrical and Information Engineering, The University of Sydney, Camperdown, NSW, AustraliaResearch Center for Augmented Intelligence, Research Institute of Artificial Intelligence, Zhejiang Laboratory, Hangzhou, ChinaThe Electroencephalogram (EEG) pattern of seizure activities is highly individual-dependent and requires experienced specialists to annotate seizure events. It is clinically time-consuming and error-prone to identify seizure activities by visually scanning EEG signals. Since EEG data are heavily under-represented, supervised learning techniques are not always practical, particularly when the data is not sufficiently labelled. Visualization of EEG data in low-dimensional feature space can ease the annotation to support subsequent supervised learning for seizure detection. Here, we leverage the benefit of both the time-frequency domain features and the Deep Boltzmann Machine (DBM) based unsupervised learning techniques to represent EEG signals in a 2-dimensional (2D) feature space. A novel unsupervised learning approach based on DBM, namely DBM_transient, is proposed by training DBM to a transient state for representing EEG signals in a 2D feature space and clustering seizure and non-seizure events visually. The effectiveness of DBM_transient is demonstrated on a widely-used benchmark dataset from Bonn University (Bonn dataset) and a raw clinical dataset from Chinese 301 Hospital (C301 dataset), with a large fisher discriminant value, surpassing the abilities of other dimensionality reduction methods, including DBM converged to an equilibrium state, Kernel Principal Component Analysis, Isometric Feature Mapping, t-distributed Stochastic Neighbour Embedding, Uniform Manifold Approximation. Such feature representation and visualization can help physicians to understand better the normal versus epileptic brain activities of each patient and thus enhance their diagnosis and treatment abilities. The significance of our approach facilitates its future usage in clinical applications.https://ieeexplore.ieee.org/document/10064189/Seizure detectionelectroencephalogram (EEG)Deep Boltzmann Machine (DBM)discrete wavelet transform (DWT)Fisher’s discriminant function |
| spellingShingle | Tengzi Liu Muhammad Zohaib Hassan Shah Xucun Yan Dongping Yang Unsupervised Feature Representation Based on Deep Boltzmann Machine for Seizure Detection IEEE Transactions on Neural Systems and Rehabilitation Engineering Seizure detection electroencephalogram (EEG) Deep Boltzmann Machine (DBM) discrete wavelet transform (DWT) Fisher’s discriminant function |
| title | Unsupervised Feature Representation Based on Deep Boltzmann Machine for Seizure Detection |
| title_full | Unsupervised Feature Representation Based on Deep Boltzmann Machine for Seizure Detection |
| title_fullStr | Unsupervised Feature Representation Based on Deep Boltzmann Machine for Seizure Detection |
| title_full_unstemmed | Unsupervised Feature Representation Based on Deep Boltzmann Machine for Seizure Detection |
| title_short | Unsupervised Feature Representation Based on Deep Boltzmann Machine for Seizure Detection |
| title_sort | unsupervised feature representation based on deep boltzmann machine for seizure detection |
| topic | Seizure detection electroencephalogram (EEG) Deep Boltzmann Machine (DBM) discrete wavelet transform (DWT) Fisher’s discriminant function |
| url | https://ieeexplore.ieee.org/document/10064189/ |
| work_keys_str_mv | AT tengziliu unsupervisedfeaturerepresentationbasedondeepboltzmannmachineforseizuredetection AT muhammadzohaibhassanshah unsupervisedfeaturerepresentationbasedondeepboltzmannmachineforseizuredetection AT xucunyan unsupervisedfeaturerepresentationbasedondeepboltzmannmachineforseizuredetection AT dongpingyang unsupervisedfeaturerepresentationbasedondeepboltzmannmachineforseizuredetection |