EEG signal processing for automated epilepsy detection
Epilepsy is regarded as one among the common neurological disorders accompanied by recurring and sudden episodes of disturbances in sensory activities of brain. Researchers are still working to discover the regions of seizure onset in human brain in order to formulate new methods for effective diagn...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
2018
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/73126 |
| _version_ | 1826121674067542016 |
|---|---|
| author | Sridharan Srividya |
| author2 | Justin Dauwels |
| author_facet | Justin Dauwels Sridharan Srividya |
| author_sort | Sridharan Srividya |
| collection | NTU |
| description | Epilepsy is regarded as one among the common neurological disorders accompanied by recurring and sudden episodes of disturbances in sensory activities of brain. Researchers are still working to discover the regions of seizure onset in human brain in order to formulate new methods for effective diagnosis and quick treatment of epilepsy. Although a lot of studies have been conducted to localize the seizures, the existing methods are found to be erroneous in many cases due to inaccurate identification of seizure onset areas. Hence, novel mathematical techniques have been developed to statistically analyze the hyper synchrony of the brain regions. Having briefed the problems, this dissertation focuses on two statistical methods to deduce regions of epileptic activities in the brain. This work subjects the EEG signals to Granger causality and Transfer Entropy analysis to identify the brain regions with high functional linkages in the epileptic patients. The entire analysis is carried out by converting the brain regions into networks using graph theoretical approach. Finally, the results are compared to validate the network formulation and localization of seizures in the brain. |
| first_indexed | 2024-10-01T05:36:03Z |
| format | Thesis |
| id | ntu-10356/73126 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T05:36:03Z |
| publishDate | 2018 |
| record_format | dspace |
| spelling | ntu-10356/731262023-07-04T15:05:52Z EEG signal processing for automated epilepsy detection Sridharan Srividya Justin Dauwels School of Electrical and Electronic Engineering DRNTU::Engineering::Electrical and electronic engineering Epilepsy is regarded as one among the common neurological disorders accompanied by recurring and sudden episodes of disturbances in sensory activities of brain. Researchers are still working to discover the regions of seizure onset in human brain in order to formulate new methods for effective diagnosis and quick treatment of epilepsy. Although a lot of studies have been conducted to localize the seizures, the existing methods are found to be erroneous in many cases due to inaccurate identification of seizure onset areas. Hence, novel mathematical techniques have been developed to statistically analyze the hyper synchrony of the brain regions. Having briefed the problems, this dissertation focuses on two statistical methods to deduce regions of epileptic activities in the brain. This work subjects the EEG signals to Granger causality and Transfer Entropy analysis to identify the brain regions with high functional linkages in the epileptic patients. The entire analysis is carried out by converting the brain regions into networks using graph theoretical approach. Finally, the results are compared to validate the network formulation and localization of seizures in the brain. Master of Science (Computer Control and Automation) 2018-01-03T06:52:48Z 2018-01-03T06:52:48Z 2018 Thesis http://hdl.handle.net/10356/73126 en 70 p. application/pdf |
| spellingShingle | DRNTU::Engineering::Electrical and electronic engineering Sridharan Srividya EEG signal processing for automated epilepsy detection |
| title | EEG signal processing for automated epilepsy detection |
| title_full | EEG signal processing for automated epilepsy detection |
| title_fullStr | EEG signal processing for automated epilepsy detection |
| title_full_unstemmed | EEG signal processing for automated epilepsy detection |
| title_short | EEG signal processing for automated epilepsy detection |
| title_sort | eeg signal processing for automated epilepsy detection |
| topic | DRNTU::Engineering::Electrical and electronic engineering |
| url | http://hdl.handle.net/10356/73126 |
| work_keys_str_mv | AT sridharansrividya eegsignalprocessingforautomatedepilepsydetection |