Saudi Cultural Missions Theses & Dissertations
Permanent URI for this communityhttps://drepo.sdl.edu.sa/handle/20.500.14154/10
Browse
3 results
Search Results
Item Restricted Improving Sleep Health with Deep Learning: Automated Classification of Sleep Stages and Detection of Sleep Disorders(Saudi Digital Library, 2024-07-07) Almutairi, Haifa; Datta, AmitavaSleep consumes roughly one-third of a person’s lifetime, and it is characterized by distinct stages within sleep cycle. The sequence of these stages at night provides insights into the quality of sleep. Poor sleep quality can have numerous consequences, including drowsiness, reduced concentration, and fatigue. Beyond sleep quality, an analysis of the sequence of sleep stages can uncover the presence of sleep disorders. This thesis aims to focus on three key research problems related to sleep. Firstly, it focuses on the classification of sleep stages using a combination of signals and deep learning models. Sleep stages are categorized into five distinct stages, namely Wake (W), non-rapid eye movement (NREM) stages comprising N1, N2, and N3, and rapid eye movement (REM) stage. Throughout the duration of sleep, individuals experience multiple cycles of sleep stages. Each cycle contains a standard allocation of each stage. An unbalanced distribution of the stages can indicate the presence of sleep disorders. Previous studies primarily classified sleep stages using a single channel of electroencephalography (EEG) signals. However, incorporating a combination of signals from electromyography (EMG) and electrooculogram (EOG) alongside EEG data provides additional features. These features extracted from muscle activity and eye movements during sleep, thereby enhancing classification accuracy. In this thesis, a robust model called SSNet is proposed to accurately classify sleep stages from a fusion of EEG, EMG, and EOG signals. This model combine convolutional neural networks (CNNs) and long short-term memory (LSTM) networks to extract the salient features from various physiological signals. The CNN architecture extracts spatial features from the input signals, while LSTM architecture captures the temporal features present in signals. This study has obtained encouraging outcomes in the classification of sleep stages through the fusion of physiological signals and deep learning techniques. Secondly, this thesis aim to detect obstructive sleep apnoea (OSA) from electrocardiography (ECG) signals using deep learning methods. Sleep disorder breathing (SDB) is categorized into three different types, which are OSA, central sleep apnoea, and mixed sleep apnoea. OSA is the most common form of SDB that is characterized by repeated interruptions in breathing during sleep, leading to fragmented sleep patterns and various health complications. Previous studies developed feature engineering methods and machine learning models for the detection of OSA. Feature engineering methods involve crafting relevant features to feed into machine learning models. However, feature engineering is time-consuming and requires domain expertise. In contrast, deep learning automatically extracts features from ECG signals for OSA detection, eliminating the need for manual feature engineering methods. In this thesis, three deep learning architectures are proposed, including standalone convolutional neural networks (CNN), CNN with long short-term memory (LSTM), and CNN with gated recurrent unit (GRU). Through rigorous experimentation and evaluation, the combination of CNN and LSTM architecture is the best-performing model for OSA detection. To further enhance the architecture’s performance, the hyperparameters of the CNN and LSTM models were tuned and tested over a large dataset to validate their effectiveness. The third research problem addressed in this thesis is detection of periodic leg movements (PLM) and SDB from NREM stage by using a combination of signals and deep learning models. PLM is characterized by involuntary leg movements during sleep. These movements can disrupt sleep and result in daytime sleepiness with reduced quality of life. Detecting PLM and SDB events during NREM stage allows for quantifying the severity of sleep disorders. Previous studies have focused on the development of signal-based models for detecting PLM or SDB. However, the models lacked the ability to distinguish these events within specific sleep stages. To address this problem, a novel deep learning architecture known as DeepSDBPLM is proposed. This architecture aims to detect PLM and SDB events during the NREM stage. This architecture incorporates novel input features called attention EMDRaw signals and utilizes a Residual Convolutional Neural Network (ResCNN) model. This thesis presents experimental results using publicly available datasets to evaluate the performance of the proposed deep learning models for classification of sleep stages, and detection of sleep disorders. The models were evaluated standard metrics. It includes accuracy, sensitivity, specificity, and F1 score. The empirical results establish the effectiveness of proposed approaches. The models can be a stepping stone towards more advanced techniques.14 0Item Restricted Advanced physiological monitoring of the cardiorespiratory system in healthy subjects and patients with sleep-disordered breathing(King's College London, 2024) Alsharifi, Abdulaziz; Steier, JoergIn this thesis, advanced physiological monitoring of the cardiorespiratory system was examined with the aim to improve outcomes for patient care in the management of respiratory conditions. The load:capacity ratio of the respiratory system was introduced as an indicator that reflects respiratory efficiency, the balance between the physiological load on the respiratory system and the capacity of the respiratory muscles. While Neural Respiratory Drive (NRD) represents a crucial component in assessing respiratory efficiency and its fundamental role in maintaining stable breathing, the load;capacity ratio provides insights into the ability to meet physiological demands, modulate symptoms, and maintain quality of sleep. Three integrated research projects examined different aspects of the load: capacity ratio of the respiratory system. The first study investigated levels of NRD, particularly the electromyographic (EMG) activity of the diaphragm, and its association with the perception of breathlessness, aiming to determine the minimally clinically important difference (MCID) in its activation. NRD, as measured by the respiratory muscle pump EMG, is a marker for inspiratory effort and closely correlates with work of breathing and symptoms, such as breathlessness. A systematic review and meta-analysis were undertaken to determine the MCID of NRD measurement using anchor-based and distribution-based methods. The systematic review identified 21 eligible studies involving 483 adult participants. NRD was primarily defined as the maximal root mean square of the diaphragm EMG as a percentage of maximum (EMGdi %max). In healthy individuals, the absolute range of EMGdi at rest was 6.0 to 16.6 %max, and 56.9 to 71.0 %max during exercise, with a mean difference of 53.3 [95% CI: 49.4; 57.1] %max. In COPD patients, the absolute range of EMGdi at rest was between 12 and 18 %max, and 59 to 72 %max during exercise, with a mean difference of 48.5 [95% CI: 44.8; 52.1] %max. The MCID of EMGdi %max, associated with a clinically large effect size, was 2.43% [95% CI: 1.96; 2.91] in normal subjects and 2.76% [95% CI: 1.92; 3.61] for COPD patients. For the Borg score, the MCID was approximately one unit for both healthy subjects and COPD patients. The second project focused on physiological experiments examining the effects of submental electrical stimulation and chemoreceptor stimulation in normal subjects during head-down tilt conditions while breathing different gas mixtures (room air, hypercapnic, and hypoxic). This research highlighted the importance of the load:capacity ratio in understanding the impact of loading the cardiorespiratory system and sensitising the baro- and chemoreceptors. Using an experimental model of baroreceptor loading induced by head-down tilt, the cardiorespiratory responses were measured during 50° head-down tilt combined with submental electrical stimulation and different gas conditions (hypercapnia, hypoxia, normoxia). The study involved 13 healthy subjects. Analysis from a three-way ANOVA indicated that blood pressure decreased significantly with transcutaneous electrical stimulation, and different gas conditions and postures similarly impacted blood pressure control. Additionally, there was increased minute ventilation with electrical stimulation, and an interaction effect between gas condition and posture on minute ventilation. The third project assessed the feasibility of remote monitoring to improve adherence to non-invasive ventilation (NIV) in patients with sleep-disordered breathing and hypercapnic respiratory failure. The study investigated remote monitoring of home mechanical ventilation (HMV) for treating chronic hypercapnic respiratory failure in patients with Obstructive Sleep Apnoea/Obesity Hypoventilation Syndrome (OSA/OHS). The primary aim was to assess whether remote monitoring could improve and optimise NIV adherence, test patients' willingness to use remote monitoring devices, and evaluate the associated healthcare resource usage and symptom improvement. In total, 32 participants were enrolled and randomly assigned to the intervention (remote monitoring) or usual care arm. The groups were similar in age, gender, and body mass index. While NIV usage in the intervention arm showed an initial improvement at the 6-week follow-up, the primary outcome of average nocturnal NIV adherence at 12 weeks was similar in both groups. Quality of life outcomes did not differ significantly between the intervention and usual care groups at the end of the trial. However, one severe adverse event with acute-on-chronic hypercapnic respiratory failure requiring hospitalization due to NIV non-adherence occurred in the control group. In conclusion, this thesis demonstrated various aspects of assessing the load:capacity ratio and the importance and complexities of understanding the underlying physiology in the context of advanced cardiorespiratory monitoring. The MCID associated with the sensation of breathlessness for the diaphragm EMG was described. A sensitization effect of electrical stimulation on the baroreceptors, impacting blood pressure, was proven. Furthermore, while the value of remote monitoring in patients with OSA/OHS requires further assessment, regular review of cardiorespiratory parameters contributes to the provision of safe follow-up surveillance in this cohort. Each project underlined the relevance of a deeper pathophysiological understanding for the management and treatment of cardiorespiratory conditions. These findings could be used to improve respiratory care by focusing on patient-based, clinically relevant outcomes and optimising healthcare delivery for clinical trials and regular clinical services for patients with respiratory conditions and sleep-disordered breathing.6 0Item Restricted Substance Use and Mental Health Conditions Among US Active Duty Military Personnel: Prevalence and Associated Factors(2023-04-13) Alulaiyan, Mohammed; Alqaderi, Hend; Tavares, Mary; Vardavas, Constantine; Alhazmi, HeshamObjective: To assess the relationship between three mental health conditions (post-traumatic stress disorder [PTSD], generalized anxiety disorder [GAD], and depressive symptoms), and cigarette smoking or marijuana use. We also explored this relationship when adding sleep duration (as a mediator variable). Methods: This was a cross-sectional study and secondary data analysis of the 2015 Department of Defense (DoD) Health Related Behaviors Survey (HRBS). Prevalence and 95% Confidence Interval (CI) of PTSD, GAD, and depressive symptoms with the sociodemographic characteristics of the United States (US) active service duty members were measured. Weighted multivariable logistic regression analyses were conducted to estimate the adjusted Odds Ratios (aOR) and 95% CI of the associations between PTSD, GAD, depressive symptoms and cigarette smoking or marijuana use. Mediation analysis was conducted to examine the role of sleep duration in the relationship between the exposures and outcomes mentioned. Results: Our study population includes 3372 service members that had at least one mental health disorder. The data showed that the prevalence of PTSD, GAD and depressive symptoms were higher among individuals who work in the Army (47.07%, 42.45%, 44.43% respectively) followed by the Navy (26.20%, 26.36%, 24.50%), Marines (15.11%, 18.90%, 20.09%), Air Force (10.08%, 10.44%, 9.52%) and lastly the Coast Guard (1.54%, 1.84%, 1.47%). The regression analyses showed that among those who did not receive mental health therapy, the estimated aOR of PTSD is 2.33 times higher for cigarette smokers compared to non-smokers [95% CI= 1.45, 3.74]. Additionally, the estimated aOR of GAD is 1.76 times higher for marijuana users compared to non-users [95% CI= 1.23, 2.51] and 2.26 times higher for cigarette smokers compared to non-smokers [95% CI= 1.60, 3.20]. Regarding depressive symptoms, the estimated aOR were higher for both marijuana users and cigarette smokers compared to non-users and non-smokers respectively. Specifically, the estimated aOR for marijuana users was 1.67 [95% CI= 1.05, 2.63], and for cigarette smokers it was 2.09 [95% CI= 1.35, 3.22]. When the association was investigated among different military branches, we found a statistically significant association between PTSD and both marijuana use and cigarette smoking among the Marines [aOR= 2.20, 95% CI= (1.18, 4.10), and aOR= 3.36, 95% CI= (1.73, 6.53)] respectively, when compared to individuals who did not use marijuana or smoke cigarettes. Among the Air Force, only marijuana use was statistically significantly associated with PTSD [aOR= 1.81, 95% CI= (1.02, 3.23)] when compared to non-users, and finally, cigarette smoking was statistically significantly associated with PTSD among Coast Guard members [aOR= 1.80, 95% CI= (1.22, 2.66)], when compared to non-smokers. For the mediation analysis, sleep duration was found to be a partial mediator in the relationship between smoking or marijuana use and mental health. Conclusion: Our study found that marijuana and cigarette use were associated with higher odds of GAD and depressive symptoms among military personnel who did not receive mental health therapy. Our findings indicated an interplay between mental health conditions, cigarette smoking, marijuana use, and sleep, and that sleep duration partially mediated the relationship, which suggest that improving sleep behavior could potentially improve mental health among individuals who smoke or use marijuana.20 0