Intelligent automated eeg artifacts handling using wavelet transform, independent component analysis and hierarchal clustering

• Billions of interconnected neurons are the building block of the human brain. For each brain activity these neurons produce electrical signals or brain waves that can be obtained by the Electroencephalogram (EEG) recording. Due to the characteristics of EEG signals, recorded signals often contaminate with undesired physiological signals other than the cerebral signal that is referred to as the EEG artifacts such as the ocular or the muscle artifacts. Therefore, identification and handling of artifacts in the EEG signals in a proper way is becoming an important research area. This paper presents an automated EEG artifacts handling approach, combining Wavelet transform, Independent Component Analysis (ICA), and Hierarchical clustering. The effectiveness of the proposed approach has been examined and observed on real EEG recording. According to the result, the proposed approach identified artifacts in the EEG signals effectively and after handling artifacts EEG signals showed acceptable considering visual inspection. © ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2017.

Ämnesord

NATURVETENSKAP  -- Data- och informationsvetenskap (hsv//swe)

NATURAL SCIENCES  -- Computer and Information Sciences (hsv//eng)

Nyckelord

Electroencephalogram (EEG)

Hierarchical clustering

Muscle artifacts

Ocular artifacts

Brain

Health care

Independent component analysis

Mobile telecommunication systems

Muscle

Wavelet transforms

Electrical signal

Electro-encephalogram (EEG)

Hier-archical clustering

Independent component analysis(ICA)

Physiological signals

Recorded signals

Visual inspection

Electroencephalography

Publikations- och innehållstyp

ref (ämneskategori)

kon (ämneskategori)