in Asleep

Asleep Develops AI Technology for Real-time Sleep Apnea Diagnosis at Home

|by Asleep

    On the 11th, Dr. Kim Jeong-whun, professor of otolaryngology at Seoul National University Bundang Hospital, and Asleep, a sleep technology company, announced that they have developed a technology that can detect sleep apnea in various noisy home environments in real time using smartphones.

    Sleep apnea is a condition where breathing is continuously interrupted during sleep, disrupting the sleep structure and quality. According to the National Health Insurance Service(NHIS), a total of 2.85 million patients received treatment for sleep apnea from 2015 to the first half of 2020.

    Dr. Kim Jeong-whun, who conducted the research, and Dr. Kim Dae-woo, Head of AI at Asleep, emphasized that "the remarkable aspect of this research is the use of an AI model trained on 22,500 sounds that can occur in homes, making it the world's first.”

    According to the joint research team from Seoul National University Bundang Hospital and Asleep, this study used more than 20,000 noise data generated in home environments, including household appliance noise, external vehicle noise, based on the data of about 1,000 sleep studies that recorded breathing sounds. Sleep studies are tests that evaluate the overall quality of sleep by tracking brainwaves, blood oxygen levels, breathing, heart rate, and eye and arm movements in a sleep lab.

    Both institutions confirmed the AI model's ability to detect sleep apnea in a home environment with noise interference. As a result, the accuracy of detecting sleep apnea was 86%.

    What makes this research unique is the ability to detect sleep apnea in real time, unlike existing sleep apnea diagnostic devices approved by the U.S. Food and Drug Administration (FDA), which require a full night's sleep to confirm the presence of sleep apnea.

    Dr. Kim Jeong-whun of Seoul National University Bundang Hospital said, "The 22,500 home environment noises used in this research encompass almost all the noises that can occur in homes," adding, "Unlike hospital environments, this technology can monitor sleep apnea in real-time in various noisy home environments." Dr. Kim Dae-woo, Head of Research at Asleep, also emphasized, "The development of an AI model trained on home environment noises is significant as it enables sleep apnea monitoring not only in hospitals but also in homes, making it a world first."

    The joint research team added, "The technology developed in this research has been applied to the application service 'Sleep Routine,' where anyone can download the app to check if they have sleep apnea."

    This research was recently published in the prestigious academic journal 'JMIR (Journal of Medical Internet Research)' in the field of health informatics.


    [Related Articles]

    2023.04.11ㅣMoney Today - Asleep Enables "Real-time Diagnosis of Sleep Apnea" at Home with AI Technology
    2023.04.11ㅣMaekyung Health - "Real-time Diagnosis" of Sleep Apnea at Home Using Smartphones
    2023.04.11ㅣSports Chosun - Using Smartphones to Diagnose Sleep Apnea "Real-time" at Home... Accuracy at 86%
    2023.04.11ㅣAsia Today - Seoul National University Bundang Hospital Develops Technology for "Real-time" Diagnosis of Sleep Apnea at Home
    2023.04.11ㅣHerald Economy - "Real-time" Diagnosis of Sleep Apnea at Home