Neuro-computational morphological model for electroencephalogram signals and artifacts analysis /
Electroencephalography (EEG) is a common and one of most widely used data collection techniques in the field of neuroscience to record human brain signals which have tremendous clinical and scientific importance. The analysis of the neural signals that is in the form of EEG recordings has assumed tr...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Subjects: | |
| Online Access: | http://studentrepo.iium.edu.my/handle/123456789/9705 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| LEADER | 058840000a22004210004500 | ||
|---|---|---|---|
| 008 | 190906s2019 my a f m 000 0 eng d | ||
| 040 | |a UIAM |b eng |e rda | ||
| 041 | |a eng | ||
| 043 | |a a-my--- | ||
| 050 | 1 | 0 | |a QP357.5 |
| 100 | 1 | |a Hamal, Abdul Qayoom, |e author |9 26039 | |
| 245 | 1 | |a Neuro-computational morphological model for electroencephalogram signals and artifacts analysis / |c by Abdul Qayoom Hamal | |
| 264 | 1 | |a Kuala Lumpur : bKulliyyah of Information and Communication Technology, International Islamic University Malaysia, |c 2019 | |
| 300 | |a xviii, 170 leaves : |b colour illustrations ; |c 30cm. | ||
| 336 | |2 rdacontent |a text | ||
| 337 | |2 rdamedia |a unmediated | ||
| 337 | |2 rdamedia |a computer | ||
| 338 | |2 rdacarrier |a volume | ||
| 338 | |2 rdacarrier |a online resource | ||
| 347 | |2 rdaft |a text file |b PDF | ||
| 500 | |a Abstracts in English and Arabic. | ||
| 500 | |a "A thesis submitted in fulfilment of the requirement for the degree of Doctor of Philosophy in Computer Science." --On title page. | ||
| 502 | |a Thesis (Ph.D)--International Islamic University Malaysia, 2019. | ||
| 504 | |a Includes bibliographical references (leaves 153-165). | ||
| 520 | |a Electroencephalography (EEG) is a common and one of most widely used data collection techniques in the field of neuroscience to record human brain signals which have tremendous clinical and scientific importance. The analysis of the neural signals that is in the form of EEG recordings has assumed tremendous importance and is an active area of research with applications varying from Brain Computer Interface (BCI) to medical applications that include diagnosis of different neural disorders like epilepsy. Analysis of EEG is also very important to better understand various human cognitive processes, human emotion, etc.While as manual analysis of EEG signals performed by EEG experts is still prevalent, many automatic analysis methods have been developed. However, there are many challenges in developing automatic analysis methods. The presence of artifacts in EEG data makes it difficult to learn the brain patterns associated with different categories of mental state. The presence of artifacts makes EEG signal difficult to interpret and may lead to serious misinterpretation. Furthermore, the presence of artifacts in EEG can lead to miscalculation in measuring important parameters that renders the diagnosis ineffective thereby reducing the efficacy as well as usefulness of EEG signals in clinical applications. In case of epileptic EEG, the seizure morphology among patients is not consistent thereby further exacerbating the analysis and as a result affecting the diagnosis.In most of BCI applications, the subject generated artifacts that contaminate the EEG signals can make the BCI communication channel ineffective as the communication mechanism should make use of neural signals only and not the artifacts. There is also scope of using EEG artifacts to control electronic equipment instead of out rightly removing the artifacts. Thus, classification of artifacts assumes much more importance irrespective of whether artifacts are to be removed at a later stage or utilised as control signals. The aim of this research is to use non-linear-based analysis model for different EEG patterns and the artifacts which can be utilised in automatic detection systems by virtue of its simplicity, effectivess and accuracy. The proposed model i.e. Neuro-computational morphological model (NCMM) is adapted from the theory of mathematical morphology. Various mathematical operations with appropriate structuring elements (SE) are used to extract the constituent signals and subsequently different features are extracted in time domain. The classification is performed using the multi-layer perceptron (MLP). The NCMM has been used to identify the EEG signals emanating from different regions of the brain. The model is able to identify normal and epileptic EEG patterns with classification accuracies reaching to 99.6% with 100% specificity and 99.4% sensitivity when identifying seizure and normal eyes-close patterns. Furthermore, artifacts data analysis is carried out that identifies whether the EEG signal pattern is due to cerebral activity or other physiological source like, eye, muscles, etc. The results for artifacts analysis show that NCMM is able to classify different artifacts with a maximum accuracy of 86.9% for eye artifact. For rest conditions of eyes-open and eyes-close the highest identification is for eyes-close (EC) with 87.4% identification in case of subject-specific scenario. However, for subject-independent identification, the average classification accuracy over all three artifacts is 58.2%. The performance of NCMM is further assessed by validating it for identification of four basic emotion states wherein the highest average accuracy is 73.9% for Happy emotion class. | ||
| 650 | 0 | |a Computational neuroscience |9 26040 | |
| 650 | 0 | |a Electroencephalography |9 26041 | |
| 650 | 0 | |a Brain |x Mathematical models |9 26042 | |
| 650 | 0 | |a Neural networks (Computer science) |9 4136 | |
| 655 | 7 | |a Theses, IIUM local | |
| 690 | |a Dissertations, Academic |x Kulliyyah of Information and Communication Technology |z IIUM |9 4793 | ||
| 710 | 2 | |a International Islamic University Malaysia. |b Kulliyyah of Information and Communication Technology |9 4794 | |
| 856 | 4 | |u http://studentrepo.iium.edu.my/handle/123456789/9705 | |
| 900 | |a sz-aaz-naw-nbm | ||
| 942 | |2 lcc |c THESIS | ||
| 999 | |c 441882 |d 472128 | ||
| 952 | |0 0 |1 0 |2 lcc |4 0 |6 T Q P 00357.00005 H00198N 02019 |7 3 |8 IIUMTHESIS |9 764541 |a IIUM |b IIUM |c THESIS |d 2022-12-09 |g 0.00 |o t QP 357.5 H198N 2019 |p 11100415015 |r 1900-01-02 |t 1 |v 0.00 |y THESIS | ||
| 952 | |0 0 |1 0 |2 lcc |4 0 |6 TS C D F QP 00357.00005 H00198N 02019 |7 3 |8 IIUMTHESIS |9 858597 |a IIUM |b IIUM |c THESIS |d 2022-12-09 |g 0.00 |o ts cdf QP 357.5 H198N 2019 |p 11100415016 |r 1900-01-02 |t 1 |v 0.00 |y THESISDIG | ||