dc.contributor.advisor | Al Nashash, Hasan | |
dc.contributor.advisor | Tariq, Usman | |
dc.contributor.advisor | Yahya, Fares | |
dc.contributor.author | El Masri, Ghinwa | |
dc.date.accessioned | 2024-02-26T08:25:57Z | |
dc.date.available | 2024-02-26T08:25:57Z | |
dc.date.issued | 2023-11 | |
dc.identifier.other | 35.232-2023.53 | |
dc.identifier.uri | http://hdl.handle.net/11073/25462 | |
dc.description | A Master of Science thesis in Biomedical Engineering by Ghenwa El Masri entitled, “Workplace Stress Management using EEG-fNIRS Data Fusion and Binaural Beat Stimulation”, submitted in November 2023. Thesis advisors are Dr. Hasan Al Nashash, Dr. Usman Tariq, and Dr. Fares Al Shargie. Soft copy is available (Thesis, Completion Certificate, Approval Signatures, and AUS Archives Consent Form). | en_US |
dc.description.abstract | Work-related stress is a common problem for employees in demanding jobs, often resulting in sleep disorders and reduced job performance. Early detection of mental stress levels in employees and implementing mitigation strategies is crucial to prevent such complications. This thesis introduces two neuroimaging modalities, Electroencephalography (EEG) and functional Near-Infrared Spectroscopy (fNIRS), which complement each other, offering insights into the shared characteristics between the brain's hemodynamic and electrical responses to stress. Alongside physiological analyses, this study incorporates behavioral, subjective, and biochemical markers. After signal processing to minimize interference and noise, Canonical Correlation Analysis (CCA) and Joint Sparse CCA (JSCCA) are employed to fuse signal features. The EEG and fNIRS features are derived from functional connectivity networks, estimated using partial directed coherence (PDC). This work's contributions include the fusion of directed connectivity between EEG and fNIRS and an exploration of the stress-mitigating potential of binaural beat stimulation (BBs) while simultaneously recording EEG and fNIRS. CCA analysis identifies the primary localization of stress in the Orbitofrontal Cortex and Dorsolateral Prefrontal Cortex (DLPFC) for the fusion between the control and stress phases of EEG and fNIRS, mostly DLPFC for the stress and mitigation phases, and frontopolar prefrontal cortex as well as DLPFC for mitigation and after mitigation phases. In terms of classification performance, CCA surpasses JSCCA in enhancing fNIRS metrics, with JSCCA being more suitable for larger datasets. CCA was reported to have an accuracy of 98.38% compared to JSCCA’s 61.91% for the Naïve Bayes Model. Overall, CCA improved Naïve Bayes classification results of control/stress classes significantly, with fNIRS classification at 78.12%, EEG classification at 77.7%. Similar improvement by CCA was observed for the classifications of stress/mitigation and mitigation/after mitigation classes as well. | en_US |
dc.description.sponsorship | College of Engineering | en_US |
dc.description.sponsorship | Multidisciplinary Programs | en_US |
dc.language.iso | en_US | en_US |
dc.relation.ispartofseries | Master of Science in Biomedical Engineering (MSBME) | en_US |
dc.subject | Occupational stress | en_US |
dc.subject | Stress mitigation | en_US |
dc.subject | Binaural beats | en_US |
dc.subject | Functional connectivity | en_US |
dc.subject | Cortisol level | en_US |
dc.subject | Functional near-infrared spectroscopy (fNIRS) | en_US |
dc.subject | Electroencephalography (EEG) | en_US |
dc.subject | Data fusion | en_US |
dc.title | Workplace Stress Management using EEG-fNIRS Data Fusion and Binaural Beat Stimulation | en_US |
dc.type | Thesis | en_US |