AI-driven platform for sleep monitoring and risk detection

Researcher:
Dr. Joachim Behar | Biomedical Engineering

The Technology

Sleep apnea (OSA) is a severe medical condition that affects millions worldwide. Besides the direct burden of OSA it is clinically shown to be related to some severe chronicle diseases like CVD, Glaucoma, diabetics, and others. Although diagnosis remains a challenge, existing home sleep tests have shown several limitations. Using simple off-the-shelf oximetry signals, the AIMlab team developed a deep-learning model to estimate the apnea-hypopnea index. Among other benefits, Clinical trials demonstrated a clear advantage over the Golden standard. A retrospective study of 12,923 polysomnography recordings from six independent databases showed superior performance amounting to a total of 115,866 hours of retrospective data. The algorithm missed 0.2% of all tests set moderate-to-severe OSA patients, against 21% for the best benchmark. The home use device can not only replace the uncomfortable high-cost sleep labs and serve as a home care device and a CDx with relevant drugs but also Recent data suggest that night-to-night variability may have an impact on an accurate diagnosis of OSA.

Advantages

  • Home use
  • Simple and accurate.

Applications and Opportunities

  • Sleep apnea
arrow Business Development Contacts
Motti Koren
Director of Business Development, Life Sciences