When you wake up, your sleep report is "automatically generated"? Smart pajamas are coming!

When you wake up, your sleep report is "automatically generated"? Smart pajamas are coming!

Compiled by: Gong Zixin

Sleep is an important part of human health, however, more than 60% of adults have poor sleep quality. Sleep behaviors such as mouth breathing, sleep apnea, and snoring are the main factors leading to poor sleep quality, and may also cause chronic diseases such as cardiovascular disease, diabetes, and depression. Therefore, effective monitoring and identification of sleep status is crucial for modern health management.

Recently, a research team led by the University of Cambridge has developed a comfortable and washable "smart pajamas" that can monitor sleep disorders such as sleep apnea at home without the use of sticky patches and bulky equipment, or going to a professional sleep clinic. The research team has developed a printed fabric sensor that can monitor breathing by detecting tiny movements of the skin even when the pajamas are worn loosely around the neck and chest.

The sensors embedded in the smart pajamas are trained with a "lightweight" artificial intelligence algorithm to identify six different sleep states with 98.6% accuracy, while ignoring normal sleep movements such as tossing and turning. The energy-efficient sensors only require a small number of sleep pattern examples to successfully identify the difference between normal and abnormal sleep.

The researchers said the smart pajamas could help people suffering from sleep disorders monitor their sleep and understand the impact of lifestyle changes on sleep. The study was published in the Proceedings of the National Academy of Sciences.

Luigi Occhipinti, a professor at the Cambridge Graphene Center and the leader of the study, said that poor sleep quality has a huge impact on physical and mental health, so proper sleep monitoring is crucial. However, the current gold standard for sleep monitoring, polysomnography (PSG), is expensive and complex, and is not suitable for long-term use at home. Home devices that are simpler than PSG, such as home sleep tests, usually only target a single condition and are bulky or uncomfortable to wear; while wearable devices such as smart watches, although more comfortable to wear, can only infer sleep quality and cannot effectively and accurately monitor sleep disorders.

“We needed a product that was comfortable and easy to use every night, but also accurate enough to provide meaningful information about sleep quality,” Occhipinti said.

Smart Pajamas for Multifunctional Sleep Behavior Monitoring System Overview To develop the smart pajamas, Occhipinti and colleagues built on their earlier work designing smart collars for people with speech impairments. The team redesigned the graphene-based sensor for breathing analysis during sleep and made several design improvements to increase sensitivity. "Thanks to our design improvements, the sensor is able to detect different sleep states while ignoring normal tossing and turning movements. The increased sensitivity also means that the smart pajamas do not need to fit snugly around the neck and can provide highly accurate readings as long as the sensor is in contact with the skin."

Sleep condition recognition model

The researchers designed a machine learning model called SleepNet that uses signals captured by sensors to identify sleep states, including nasal breathing, mouth breathing, snoring, teeth grinding, central sleep apnea and obstructive sleep apnea. SleepNet is a "lightweight" artificial intelligence network that reduces computational complexity to the point where it can run on a portable device without the need to connect to a computer or server.

"We streamlined the AI ​​model to achieve the highest accuracy while achieving the lowest computational cost," Occhipinti said. "This allowed us to embed the main data processor directly into the sensor." The researchers tested the smart pajamas on healthy people and patients with sleep apnea, and were able to detect a range of sleep states with 98.6% accuracy. By giving the smart pajamas a special starching treatment, the researchers increased the durability of the sensors, allowing them to be washed in a regular washing machine. The latest version of the smart pajamas also has wireless data transmission capabilities, which can securely transmit sleep data to a smartphone or computer.

Features of the device

Sleep is essential to health, and reliable sleep monitoring is key to preventive care. Occhipinti pointed out that this smart clothing can be used at home, and it can remind users to pay attention to changes in their sleep and then discuss them with their doctors. Sleep behaviors such as nasal breathing and mouth breathing are not usually paid attention to in sleep analysis in the medical service system, but they may be an indicator of sleep disorders.

The researchers hope the sensor could be adapted for a variety of health conditions or for home uses, such as baby monitoring. They are also working to make the sensor more durable for long-term use.

Note: The cover image is a copyrighted image. Reprinting it may cause copyright disputes.

<<:  In "Nezha 2", why did General Octopus dare to roast his own "feet" to eat?

>>:  Are green beans not allowed at dinner parties? Beware: Eating this vegetable undercooked can cause poisoning

Recommend

Tips for building personal private domain traffic

The concept of private domain traffic has been ta...

What does the universe smell like?

Imagine if you had the opportunity to go into spa...

I really want to ask: How many eggs is the healthiest to eat in a day?

There is a super ingredient that is available in ...

Today, unlock another beauty of Xinjiang

If returning to the ocean is the ultimate fate of...

How can a product achieve explosive growth in users?

Competition among Internet products is essentiall...

Tips for effective advertising!

Everyone who has used competitive advertising kno...

Locusts are so hateful, can they be exterminated?

Produced by: Science Popularization China Author:...

How to Evolve Neural Networks with Automatic Machine Learning

For most people working in machine learning, desi...

3 basic elements of community operation and 11 cases

Community is an area that almost all Internet pro...

4-step method for product data operation

Product managers do a lot of things from 0 to 1, ...

It’s 40℃ now. Will it get hotter? The answer may make you “crash”…

Recently, many parts of the country have seen hig...

Makeup tips for beginners

Introduction to makeup skills resources suitable ...

A growth activity idea and plan

For Internet people who are engaged in user growt...