Researchers from the University of Cambridge have developed a groundbreaking set of smart pajamas designed to monitor breathing patterns during sleep and detect disorders such as sleep apnea, snoring, and bruxism in home environments. This innovation is made possible by advanced graphene-based strain sensors embedded directly into the fabric. The technology offers a patient-friendly alternative to conventional hospital-based sleep diagnostics, which are often cumbersome and uncomfortable.
The smart pajamas were developed by a research team at the Cambridge Graphene Centre, led by Professor Luigi Occhipinti. The team drew on their previous work involving smart chokers for individuals with speech impairments. For this project, the sensors were redesigned and enhanced for greater sensitivity and integrated into the collar of the pajamas.
“The throat area is incredibly rich in data. In our earlier project, we developed a sensor embedded in smart fabric to monitor speech signals, which differ from breathing patterns in terms of amplitude and frequency,” explains Prof. Luigi Occhipinti, Director of Research in Smart Electronics, Biosystems and AI at Cambridge Graphene Centre. “This allowed us to gather meaningful diagnostic data from patients with speech impairments—such as those recovering from laryngectomy or stroke-induced dysarthria. It’s a clear example of how much can be learned from this part of the body.”
The pajamas are intended as a more comfortable alternative to polysomnography, the gold-standard method for diagnosing sleep disorders, which requires patients to spend a night in a hospital or sleep clinic connected to numerous sensors and wires.
“Our goal was to enable patients to monitor their health in their own homes, in a natural environment,” Prof. Occhipinti says. “So, we developed a comfortable, durable sleepwear-based monitoring system resistant to movement artifacts—such as those caused by tossing and turning during sleep.”
The core technology is a graphene strain sensor (or an array of such sensors) capable of detecting minute fabric tension changes with remarkable sensitivity—even when the shirt is loosely worn around the neck.
“With our specific sensor design and materials, we can detect up to six distinct sleep-related patterns—such as nasal or mouth breathing, bruxism (teeth grinding or jaw clenching), snoring, and crucially, both central and obstructive sleep apnea,” the professor notes.
To analyze the sensor data, the team developed a machine learning model named SleepNet. It uses input from the graphene sensors to identify different sleep states, including nasal breathing, oral breathing, snoring, teeth grinding, and the two types of sleep apnea. Importantly, SleepNet functions without requiring a connection to external computers or servers. The latest versions of the pajamas can wirelessly transmit results to smartphones or computers.
“SleepNet processes and learns from both healthy individuals and those with sleep disorders. It classifies signal patterns and associates them with specific sleep conditions,” says Prof. Occhipinti. “The algorithm also filters out most noise to enable real-time preprocessing of data at the edge—on the device itself. It can transmit results for AI model training and be adapted for new patients or conditions. It’s a lightweight AI network with minimal computational complexity, yet highly accurate, enabling real-time operation.”
The smart pajamas have been tested on both healthy volunteers and individuals with sleep apnea. They demonstrated the ability to detect various sleep phases with 98.6% accuracy. Researchers hope to further adapt the sensors for other medical conditions beyond sleep monitoring and are currently working on improving their durability, including the ability to withstand repeated washes in a standard washing machine.
The importance of detecting and treating sleep disorders cannot be overstated. Proper sleep hygiene is essential for both mental and physical health. Poor sleep quality is linked to chronic conditions such as cardiovascular disease, diabetes, and depression. According to the 2025 report “The Sleep of Poles: Economic Losses,” prepared by UCE Research and the ePsycholodzy.pl platform, over 41% of Polish adults rate their sleep quality as poor.
This innovation from Cambridge could soon offer a highly accurate, comfortable, and accessible tool to combat this widespread issue—right from the comfort of home.
