Theories and Applications of Non-Contact Sleep Monitoring using Microwave Doppler Radar
- Publication Type:
- Thesis
- Issue Date:
- 2020
Open Access
Copyright Clearance Process
- Recently Added
- In Progress
- Open Access
This item is open access.
Obstructive sleep apnea (OSA) is a common and potentially lethal sleep disorder affecting at least 4% of adult males and 2% of adult females worldwide. Early detection, treatment and continuous monitoring of OSA are extremely important as it may reduce the risks associated with cardiovascular comorbidities. Polysomnography (PSG) is the gold-standard to diagnose OSA, however there are limitations, such as its unsuitability for long-term continuous monitoring.
The Thesis is a response to the demands for the non-contact sleep monitoring systems. The demands arise due to the limitations of the PSG system, the importance of early screening for OSA, the need for long-term continuous monitoring and the concern with respect to patient discomfort when using the gold-standard PSG system. The research presented in the Thesis are the novel theories, real-life applications and the results of the non-contact sleep monitoring using the non-contact microwave Doppler radar, including the “non-stationary” and “non-direct facing” subjects’ measurements in the complex sleep environment.
The novel theories that the Thesis contributes to the field of non-contact sleep monitoring are:
1. Relative Demodulation – a novel theory and technique for real-time demodulation of the subject’s chest or abdomen periodic motions using non-contact microwave Doppler radar.
2. Pulmonary Ventilation Mathematical Model – a novel mathematical model of the physiological pulmonary ventilation that enables the estimation of tidal volume using non-contact microwave Doppler radar.
3. External Ventilation Mathematical Model – a novel mathematical model of the physiological external ventilation that enables the estimation of oxygen saturation using non-contact microwave Doppler radar.
4. 3-Dimensional Feature Representation and Extraction – a novel theory and technique that represents and extracts features in 3-dimensional space. This technique, when combine with the artificial neural networks (ANN) will enable the predictions of body orientations and oxygen saturation using non-contact microwave Doppler radar.
The novel non-contact sleep monitoring real-life applications and results that the Thesis contributes to the field of non-contact sleep monitoring are:
1. Respiratory rate – achieves 91.53% accuracy with median error of ±1.30 breaths/min.
2. Heart rate – achieves 91.28% accuracy with median error of ±6.20 beats/min.
3. Tidal volume – achieves 83.13% accuracy with median error of 57.32 milliliters.
4. Body orientations – achieve high correct classification rate of 99.9%. The misclassification is at a negligible rate of 0.1%.
5. Oxygen saturation – achieves correlation coefficient of 0.92 and the 95% limits of agreement is ±2.7 (% oxygen saturation).
The contributions of the novel theories, real-life applications and the results presented in the Thesis demonstrated a good level of accuracies. The potential applications include non-contact sleep early screening and/or continuous monitoring of the respiratory and heart rates, tidal volume, body orientations and saturation oxygen during sleep. This can be use in homes, hospitals, primary care sectors, nursing home facilities and/or sleep laboratories.
Please use this identifier to cite or link to this item: