Use this URL to cite or link to this record in EThOS:
Title: Investigation of ocular artefacts in the human eeg and their removal by a microprocessor-based instrument
Author: Ifeachor, Emmanuel Chigozie
ISNI:       0000 0001 3586 1562
Awarding Body: Plymouth Polytechnic
Current Institution: University of Plymouth
Date of Award: 1984
Availability of Full Text:
Access from EThOS:
Access from Institution:
The Electroencephalogram (EEG) is widely used in clinical and psychological situations, but it is often seriously obscured by ocular artefacts (OAs) resulting from movements in the ocular system (eyeball, eyelids etc). 'The work described in this thesis is concerned with the problems of OAs in the human EEG, their removal both off-line and on-line, and the design and development of an on-line OA removal system, together with a critical review of the literature on the subject. The work of Jervis and his co-workers was extended to further study OAs, to obtain improved measures of the effectiveness of OA removal, and to find the most effective model for removing OA on-line. A number of criteria were devised to compare the performance of several models, including a more reliable pictorial method. It was found unnecessary to use the vertical and horizontal EOGs for both eyes (ie. four EOGs) in a removal model, as previously reported. This was shown to be due to strong correlation between the EOGs. It was shown that the assumption of uncorrelated error terms, implicit in present removal models, is invalid. To remedy this, the error terms were modelled as an autoregressive series. New on-line removal algorithms based on numerically stable factorization algorithms were developed. Compared to the present on-line methods the algorithms are superior, requiring no subjective manual adjustments, or the co-operation of subjects which cannot always be guarranteed. The algorithms were shown to give similar results to their off-line equivalents. A simpler algorithm based on the present on-line method is also proposed as an alternative, but may lead to a reduced performance. An important part of this research lay in the application of the results to the design and development of a new automatic OA removal system utilizing the algorithms described above.
Supervisor: Not available Sponsor: Department of Neurological Sciences, Freedom Fields Hospital, Plymouth
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Medicine