MSc thesis project proposal

[2020-21, in contact with Philips Research] Contactless Vital Signs Monitoring via Radar Sensing for Sleep Applications

CONTEXT: Despite sleep apnea being one of the most common sleep disorders, it often remains undiagnosed, or even unnoticed, due to both unawareness of its symptoms and the flaws and complexity of current diagnostic methodologies. Radar is a promising technology to monitor a sub-set of the diagnostic parameters contactless, including respiration and heart rate. A reliable contactless alternative would not only improve patient comfort, but it could also make sleep monitoring applicable outside the clinic in home environments.

There are three types of apnea: central, obstructive and mixed. During central apnea the brain fails to signal the muscles to breathe, leading to an absence of both respiratory effort and (air) flow. Obstructive sleep apnea (OSA, 80 percent of all cases) is the most common type of sleep apnea where the effort is still present but there is no flow because of the collapse of the upper airway, preventing air inflow.
During an OSA event, the respiratory movements of the chest and abdomen often change from in-phase to counter-phase, as visualized in Fig. 1. In order to detect and classify such events, it is important that the radar can monitor the movements of the chest and abdomen independently and continuously, as state of the art medical devices can do.
Besides sleep apnea, features extracted from respiration and heart rate have shown their promise in the classification of sleep stages. Therefore, radar also has its potential as a contactless sleep monitor. As the information is predominantly present in the variability of the signals, beat-to-beat/breath-to-breath accuracy rather than average heart/breathing rate is required.

Assignment

1] Investigate how chest and abdominal movements can be monitored independently in a sleep setting using radar(s) where the patients can be in either supine, side or prone position. Here the radar measurements should be compared to the signals measured by the respiratory inductance plethysmography (RIP) bands attached to the chest and abdomen.
2] Investigate the feasibility of radar-based heart rate monitoring with beat-to-beat accuracy for sleep apnea detection and sleep stage classification. To assess the accuracy, the radar-based measurement should be compared to the ground-truth ECG.

This initial investigation should aim at demonstrating feasibility in healthy subjects (“proof-of-concept”) in a simulated sleep environment. Experiments on actual patients are beyond the scope of this assignment.
The student is expected to perform a combination of experimental and signal processing work, contributing to the design and validation of a setup to collect relevant radar data, as well as the development of signal processing algorithms that can provide the required details of thorax/abdomen movements and breath-to-breath / heart beat-to-beat accuracy.
Once a reliable and effective framework for radar-based monitoring is established, the combination of information from the radar sensors and cameras in a multimodal fusion framework can also be considered within the scope of this project, combining the expertise of the MS3 group at TUD in radar processing and Philips’ expertise on image processing. In first instance, the camera could provide an estimate of the position and orientation of the subject to enhance the performance of the radar estimation (localization of the area of maximum movement, adaptation of the signal processing depending on the person’s position).
The student is expected to spend the majority of his time at TU Delft using the facilities of the MS3 research group, but attending periodic meetings involving the team at Philips Research Eindhoven. A placement at the Philips facilities in Eindhoven can also be considered in due course, depending on progress and needs within the project workload.

Contact

dr. Francesco Fioranelli

Microwave Sensing, Signals and Systems Group

Department of Microelectronics

Last modified: 2021-01-19