Catch More Z’s with These High-Tech PJs

Getting a poor night’s sleep will leave you dragging all day, and that is not just a matter of comfort. Chronic sleep issues can lead to serious conditions such as cardiovascular disease, diabetes, and depression. Loss of sleep also results in serious economic consequences — it has been estimated that it leads to a one percent reduction in global GDP and dozens of lost days of work each year for every worker.

Fortunately, many sleep disorders can be addressed through some type of medical intervention. But before that can happen, an accurate diagnosis must be made. And that is one reason why untreated sleep disorders are so prevalent — traditional diagnostic tools are either cumbersome, expensive, or minimally effective.

Polysomnography is the gold standard test, but it is expensive, uncomfortable, and must be carried out in a specialized lab, which prevents many from seeking this form of testing. At home tests and smart devices can only measure a few parameters, and are not especially accurate, so they are of limited value and may make inaccurate diagnoses.

An accurate, inexpensive, and comfortable test that can be taken at home is sorely needed to make some headway in the fight against sleep disorders. And that is exactly what a team led by researchers at the University of Cambridge has just proposed. They have developed smart pajamas that can conduct a comprehensive sleep study in the comfort of one’s own home.

These smart pajamas are embedded with highly sensitive printed fabric sensors that detect even the tiniest movements in the skin, allowing them to monitor breathing patterns in real-time. Unlike traditional sleep tests that require numerous sensors and wires, these pajamas can be worn loosely around the neck and chest while still providing accurate readings.

The secret behind this technology lies in its graphene-based sensors and an advanced lightweight artificial intelligence (AI) algorithm called SleepNet. In conjunction with SleepNet, the sensors detect six different sleep states, including nasal breathing, mouth breathing, snoring, teeth grinding, central sleep apnea, and obstructive sleep apnea. The team also introduced a starching step to improve the durability of the sensors, allowing the pajamas to be machine-washable without compromising performance.

While traditional AI models often require vast amounts of computing power, SleepNet has been optimized to run on small, portable devices. The team managed to prune the AI model to reduce computational demands while maintaining high accuracy, allowing it to function independently without needing to connect to external servers or computers. This not only allows the system to run on a low-power device at home, but it also keeps the collected data private.

To test the pajamas, the researchers conducted trials on both healthy individuals and patients with diagnosed sleep apnea. The results showed that the smart garment could detect a range of sleep states with near-perfect accuracy. Overall, SleepNet has been shown to be capable of achieving an impressive 98.6% accuracy level in distinguishing between normal sleep and disordered breathing.

Beyond sleep monitoring, the research team envisions broader applications for this technology. Discussions are already underway to adapt the sensors for baby monitoring, helping parents track their child's breathing and sleep patterns. The team is also working on further improving the longevity and resilience of the sensors to make them suitable for long-term use. With this additional work, these smart pajamas could help millions achieve better sleep — and better health.