Classification of labour pain using electroencephalogram signal based on wavelet method / Sai Chong Yeh

Sai , Chong Yeh (2020) Classification of labour pain using electroencephalogram signal based on wavelet method / Sai Chong Yeh. PhD thesis, Universiti Malaya.

[img] PDF (The Candidate's Agreement)
Restricted to Repository staff only

Download (212Kb)
    PDF (Thesis PhD)
    Download (2787Kb) | Preview


      Electroencephalogram (EEG) is the recording of electrical activity of the cerebral cortex through electrodes placed on the scalp. EEG is used to acquire neurophysiological signals for application in clinical diagnosis and brain computer interface (BCI). However, in practical settings the EEG signals are often contaminated by signal artifacts known as the biological and environmental artifacts. These artifacts degrade EEG signals, thereby obstructing clinical diagnosis or BCI applications by distorting the observed power spectrum. Procedures for automated removal of EEG artifacts are frequently sought after in pre-processing and filtering of the EEG signals. In recent years, a combination of independent component analysis (ICA) and discrete wavelet transform (DWT) has been introduced as standard technique for EEG artifact removal. However, in performing the wavelet-ICA procedure, visual inspection or arbitrary thresholding may be required to identify the artifactual components in the EEG signal. This study proposed an integrated system for EEG signals pre-processing by using machine learning algorithms in the identification of artifactual components during the process of Wavelet-ICA. Supervised and unsupervised machine learning algorithms particularly the Support Vector Machine (SVM) and Density Based Spatial Clustering of Application with Noise (DBSCAN) are used in this study. These methods present a robust system that enables fully automated identification and removal of artifacts from EEG signals, without the need of visual inspection or arbitrary thresholding. The training and parameters selection of the machine learning algorithms are conducted using EEG data collected from ten subjects in the laboratory. Using test data contaminated by eye blink artifacts and public dataset from EEGLAB, it was shown that these methods performed better in identifying artifactual components than did existing thresholding methods. Furthermore, wavelet-ICA in conjunction with machine learning algorithm successfully removed target artifacts, while largely retaining the EEG source signals of interest. This method is also extendable to accommodate multiple types of artifacts present in multichannel EEG. As a practical application of this study, the developed system is used in an application to monitor pain response due to uterine contractions during labour. This part of the study aimed to assess the utility of EEG as an objective marker of pain during the first stage of labour. We obtained EEG and cardiotocography (CTG) data in ten parturient women during their first stage of labour. The study subjects reported the extent of their pain experienced due to uterine contractions, which were recorded by the CTG tracing. Simultaneous 16-channels EEG traces were obtained for spectral analysis and a subsequent classification using SVM aiming to predict the pain experienced in relation to uterine contractions. It was found that pain due to uterine contraction correlated positively with relative delta and beta band activities and negatively with relative theta and alpha band activities of the EEG signals. SVM using the spectral activities, statistical and non-linear features classified the state of pain with an accuracy of 83% using a classification model generalizable across subjects. Furthermore, dimension reduction using principal component analysis (PCA) successfully reduced the number of features used in the classification while achieving a maximum classification accuracy of 84%. The results shown that continuous EEG affords the means to assess objectively maternal pain experienced. All in all, this study aims to design, develop, optimize and test the method of pain assessment using the EEG signal during the active contraction phase of the first stage of labour. Future studies are envisioned to investigate EEG markers of pain in other clinical states, aiming to generalize the use of EEG as an objective method of pain assessment.

      Item Type: Thesis (PhD)
      Additional Information: Thesis (PhD) - Faculty of Engineering, Universiti Malaya, 2020.
      Uncontrolled Keywords: Electroencephalogram; Machine learning; Pain assessment; Wavelet-ICA; Support Vector Machine (SVM)
      Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
      Divisions: Faculty of Engineering
      Depositing User: Mr Mohd Safri Tahir
      Date Deposited: 29 Nov 2021 02:23
      Last Modified: 16 Jan 2023 06:56

      Actions (For repository staff only : Login required)

      View Item