Artefact removal for neonatal electroencephalogram
Neonatal electroencephalogram (EEG) provides vital diagnostic/prognostic insight. Apartnfrom vigilant state (e.g. induced by seizure, Central Nervous System diseases etc.) detection, it is also the cornerstone in Sleep-Wake-Cycle (SWC) recognition. SWC recognition holds great clinical significance a...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2013
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/54471 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Neonatal electroencephalogram (EEG) provides vital diagnostic/prognostic insight. Apartnfrom vigilant state (e.g. induced by seizure, Central Nervous System diseases etc.) detection, it is also the cornerstone in Sleep-Wake-Cycle (SWC) recognition. SWC recognition holds great clinical significance as abnormalities in SWC are often indicators of neurological diseases. To facilitate SWC recognition, artefact removal is often great importance since neonatal EEG is often contaminated with all kinds of artefacts. High amplitude and high frequency (HAHF) artefacts are a major source of distortion in neonatal EEG, which greatly impedes further processing. This thesis proposes an approach that combines wavelet decomposition and principal component analysis, for the removal of HAHF artefacts in neonatal EEG. The suggested approach demonstrates superior performance as compared with Elliptic low pass filter and median filter. |
|---|