Automated Seizure Detection Based on State-Space Model Identification
In this study, we developed a machine learning model for automated seizure detection using system identification techniques on EEG recordings. System identification builds mathematical models from a time series signal and uses a small number of parameters to represent the entirety of time domain sig...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/24/6/1902 |
| _version_ | 1827305036411043840 |
|---|---|
| author | Zhuo Wang Michael R. Sperling Dale Wyeth Allon Guez |
| author_facet | Zhuo Wang Michael R. Sperling Dale Wyeth Allon Guez |
| author_sort | Zhuo Wang |
| collection | DOAJ |
| description | In this study, we developed a machine learning model for automated seizure detection using system identification techniques on EEG recordings. System identification builds mathematical models from a time series signal and uses a small number of parameters to represent the entirety of time domain signal epochs. Such parameters were used as features for the classifiers in our study. We analyzed 69 seizure and 55 non-seizure recordings and an additional 10 continuous recordings from Thomas Jefferson University Hospital, alongside a larger dataset from the CHB-MIT database. By dividing EEGs into epochs (1 s, 2 s, 5 s, and 10 s) and employing fifth-order state-space dynamic systems for feature extraction, we tested various classifiers, with the decision tree and 1 s epochs achieving the highest performance: 96.0% accuracy, 92.7% sensitivity, and 97.6% specificity based on the Jefferson dataset. Moreover, as the epoch length increased, the accuracy dropped to 94.9%, with a decrease in sensitivity to 91.5% and specificity to 96.7%. Accuracy for the CHB-MIT dataset was 94.1%, with 87.6% sensitivity and 97.5% specificity. The subject-specific cases showed improved results, with an average of 98.3% accuracy, 97.4% sensitivity, and 98.4% specificity. The average false detection rate per hour was 0.5 ± 0.28 in the 10 continuous EEG recordings. This study suggests that using a system identification technique, specifically, state-space modeling, combined with machine learning classifiers, such as decision trees, is an effective and efficient approach to automated seizure detection. |
| first_indexed | 2024-04-24T17:49:17Z |
| format | Article |
| id | doaj.art-466c55b82ba041f49f809dc5e906eade |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-24T17:49:17Z |
| publishDate | 2024-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-466c55b82ba041f49f809dc5e906eade2024-03-27T14:04:06ZengMDPI AGSensors1424-82202024-03-01246190210.3390/s24061902Automated Seizure Detection Based on State-Space Model IdentificationZhuo Wang0Michael R. Sperling1Dale Wyeth2Allon Guez3Department of Electrical and Computer Engineering, Drexel University, Philadelphia, PA 19104, USADepartment of Neurology, Thomas Jefferson University, Philadelphia, PA 19107, USAJefferson Hospital for Neuroscience, Philadelphia, PA 19107, USADepartment of Electrical and Computer Engineering, Drexel University, Philadelphia, PA 19104, USAIn this study, we developed a machine learning model for automated seizure detection using system identification techniques on EEG recordings. System identification builds mathematical models from a time series signal and uses a small number of parameters to represent the entirety of time domain signal epochs. Such parameters were used as features for the classifiers in our study. We analyzed 69 seizure and 55 non-seizure recordings and an additional 10 continuous recordings from Thomas Jefferson University Hospital, alongside a larger dataset from the CHB-MIT database. By dividing EEGs into epochs (1 s, 2 s, 5 s, and 10 s) and employing fifth-order state-space dynamic systems for feature extraction, we tested various classifiers, with the decision tree and 1 s epochs achieving the highest performance: 96.0% accuracy, 92.7% sensitivity, and 97.6% specificity based on the Jefferson dataset. Moreover, as the epoch length increased, the accuracy dropped to 94.9%, with a decrease in sensitivity to 91.5% and specificity to 96.7%. Accuracy for the CHB-MIT dataset was 94.1%, with 87.6% sensitivity and 97.5% specificity. The subject-specific cases showed improved results, with an average of 98.3% accuracy, 97.4% sensitivity, and 98.4% specificity. The average false detection rate per hour was 0.5 ± 0.28 in the 10 continuous EEG recordings. This study suggests that using a system identification technique, specifically, state-space modeling, combined with machine learning classifiers, such as decision trees, is an effective and efficient approach to automated seizure detection.https://www.mdpi.com/1424-8220/24/6/1902EEGsystem identificationstate-space modelautomated seizure detection |
| spellingShingle | Zhuo Wang Michael R. Sperling Dale Wyeth Allon Guez Automated Seizure Detection Based on State-Space Model Identification Sensors EEG system identification state-space model automated seizure detection |
| title | Automated Seizure Detection Based on State-Space Model Identification |
| title_full | Automated Seizure Detection Based on State-Space Model Identification |
| title_fullStr | Automated Seizure Detection Based on State-Space Model Identification |
| title_full_unstemmed | Automated Seizure Detection Based on State-Space Model Identification |
| title_short | Automated Seizure Detection Based on State-Space Model Identification |
| title_sort | automated seizure detection based on state space model identification |
| topic | EEG system identification state-space model automated seizure detection |
| url | https://www.mdpi.com/1424-8220/24/6/1902 |
| work_keys_str_mv | AT zhuowang automatedseizuredetectionbasedonstatespacemodelidentification AT michaelrsperling automatedseizuredetectionbasedonstatespacemodelidentification AT dalewyeth automatedseizuredetectionbasedonstatespacemodelidentification AT allonguez automatedseizuredetectionbasedonstatespacemodelidentification |