The following technology is a set of algorithms and a system that enables the detection of sleep quality,
snoring severity and Obstructive Sleep Apnea (OSA) using audio signals acquired by a non-contact
microphone.
The Clinical Need
Statistics on snoring are often contradictory, but at least 30% of adults snore. Though snoring is often
considered a minor affliction, snorers can sometimes suffer severe impairment of lifestyle and other
health indications such as development of carotid artery atherosclerosis, the risk of brain damage, stroke
and OSA. OSA is a chronic disorder that can lead to considerable morbidity. Partial or complete collapse
of the upper airway during sleep has different effects on the human body, ranging from noisy breathing
(simple snoring) to cardiovascular morbidity. Globally, >200 million people are thought to have the
condition, with an overwhelming 90% of them remaining undiagnosed. Adequate diagnosis and treatment
of OSA reduces complications and improves the quality of life.
Development Stage and Development Status-Summary
A system that analyses the nocturnal audio signal and combines several developed acoustic features
with a classifier was developed. Development stages already achieved include: a) Breathing sounds and
snore detection and analysis algorithm, b) Sleep/wake detection algorithm, c) OSA detection and severity
estimation algorithms, d) Audio database of patients from sleep lab and at-home recordings. The results
that were achieved in clinical studies using a combined algorithm were very good: a) Breathing sounds
and snore detection – above 98% detection rate, b) Sleep/wake detection – show excellent diagnostic
agreement according to several parameters, epoch-by-epoch detection rate above 84%, c) OSA severity
estimation – correlation of 0.89 with the gold-standard at laboratory sleep study.
Goals and Benefits
Currently, polysomnography (PSG) is the gold standard approach for sleep evaluation (including sleep
quality estimation and OSA diagnosis). The market share of home testing devices and cellular health
applications related to sleep disorders is expected to increase significantly, as they are cost-effective, and
more convenient, as is the case with sound detection based devices. Our system diagnostic's quality is
comparable to home testing devices approved for medical use for OSA and other sleep disorder
diagnosis. Our system enables the analysis of variety of important medical and physiological parameters,
such as total-sleep-time, snore detection, snore severity estimation, and sleep/wake patterns to
determine sleep quality and OSA.
Potential Commercial Uses and Market
The technology can be used as a device/cellular application for home detection of sleep quality, snoring
severity and OSA. It can be used to evaluate the effectiveness of snoring relieves, OSA treatments and
lifestyle modifications related to sleep disorders. A potential business model may be based on
collaboration with snoring and therapeutics producers for a bundled deal of a therapeutic device and
evaluation application, or commercialize independent free online application, generating revenues from
users direction toward collaborating therapeutic producers or medical diagnostic services.
The worldwide OSA (Diagnostic & Therapeutic) devices market is expected to reach $5.3 Billion by 2020
from an estimated $3.7 Billion in 2015.
Research Team
Dr. Zigel Yaniv, Department of Biomedical Engineering, Ben-Gurion University of the Negev, Beer-Sheva,
Israel; Prof. Ariel Tarasiuk, Sleep-Wake Disorder Unit, Soroka University Medical Center, Department of
Physiology, Ben-Gurion University of the Negev, Beer-Sheva, Israel.