University of Chester researchers developed a deep learning model that monitors physiological markers to address diagnostic delays and costs.

Key takeaways:

University of Chester researchers developed a wearable system achieving over 95% precision in sleep apnea detection.
The device monitors respiratory activity, heart rate, blood oxygen saturation, and body posture in real time.
The system architecture includes AI model deployment, a mobile application, and cloud-based storage.
Researchers aim to provide a scalable alternative to costly in-lab polysomnography and slow diagnostic pathways.

A new study from the University of Chester presents a wearable artificial intelligence (AI)-driven system designed to detect sleep apnea in real time with high precision.

The research, published as a dissertation titled A Wearable AI-Driven System for Real-Time Detection of Sleep Apnoea, details a device equipped with multiple sensors. These sensors continuously monitor respiratory activity, heart rate, blood oxygen saturation levels, and body posture.

According to the dissertation, the deep learning-trained model achieved an apnea detection precision of over 95%. The system provides real-time results, enabling timely visual feedback and the potential activation of therapeutic stimulators to address apnea events.

The system architecture integrates AI model deployment, mobile application development, and cloud-based storage infrastructure. This design supports continuous monitoring, model updates, and remote analysis.

The project was authored by Yurui Zheng, PhD (pictured), an electronic and electrical engineering research degree graduate, and advised by professor Bin Yang and associate professor Theo Papadopoulos. It was conducted in collaboration with PFL Healthcare.

“Sleep apnea remains underdiagnosed due to the high cost and slow access to traditional testing,” says Zheng in a release. “Some advanced home diagnostics cost around £348. Clinical in-lab PSG starts at £880 and takes roughly 10 days. NHS pathways are often slower, taking months and multiple appointments.”

The research aims to offer a scalable, cost-effective solution for non-invasive monitoring, bridging the gap between clinical diagnostics and home-based sleep health management.

“The vision for wearable AI in sleep care is to create a seamless, noninvasive solution that continuously monitors sleep patterns, detects anomalies, and even intervenes to improve sleep quality,” says Zheng in a release. “I’m excited about the potential to translate cutting-edge AI into tangible health benefits—empowering people to manage sleep disorders proactively, anytime, anywhere.”

The researchers highlight the project’s ability to apply engineering solutions to healthcare needs.

“This research highlights how engineering and AI can directly address real-world healthcare challenges,” says Papadopoulos in a release. “Congratulations to Dr Zheng and Prof Yang on developing a scalable, patient-centred, and cost-effective solution that bridges the gap between the clinical laboratory and everyday well-being.”