Israeli Scientists Invent Innovative Method for Diagnosing Sleep Apnea

In Israel, a technology based on artificial intelligence has been developed that accurately identifies all significant cases of moderate and severe obstructive sleep apnea syndrome. It is expected to be implemented as an accessible mobile application that will allow individuals to self-diagnose this type of sleep disorder, which poses a serious health threat, and seek medical assistance.

Sleep apnea is a major problem. Not only because the person suffering from it disturbs their loved ones' sleep with loud snoring. It is characterized by temporary pauses or weakening of breathing during sleep, occurring every few minutes and lasting more than ten seconds. Oxygen intake into the body decreases, resulting in damage to various cells, primarily heart and brain cells. The risk of developing diabetes and cardiovascular diseases increases, which can ultimately lead to a heart attack or stroke.

As a result of stress-related changes, sleep becomes superficial and does not provide full rest. During the day, a person suffering from sleep apnea often experiences severe drowsiness, irritability, and difficulties concentrating, leading to decreased work performance.

Breathing disorders during sleep are usually found in adulthood – more often in men, people with significant overweight, and smokers. According to the World Health Organization, about 20% of the global population lives with apnea. This condition is successfully treated today; however, many people are unaware of their diagnosis, which prevents them from receiving necessary medical care.

An international team of researchers led by scientists from the Technion – Israel Institute of Technology has taken on this problem. They conducted extensive work, resulting in a technology based on artificial intelligence that accurately diagnoses obstructive sleep apnea syndrome. The details of the project and the main findings were shared by the scientists in the latest issue of the scientific journal EClinicalMedicine.

In creating the innovative technology developed by leading Technion bioengineering specialist Joachim Behar and his colleagues, data from nearly 900 adult residents of São Paulo, Brazil, were used. After processing this data, the artificial intelligence system successfully identified individuals with and without sleep apnea.

The diagnosis was made after summarizing the data obtained from the experiment participants, which fell into the category of biological markers for sleep apnea, including blood oxygen saturation levels during sleep, age, height, weight, and neck circumference.

Currently, there are two main methods for diagnosing obstructive sleep apnea syndrome. The first is polysomnography, during which specialists use monitoring equipment to observe the body's activity during sleep, determining the frequency and duration of breathing pauses. This is conducted in specialized sleep laboratories. The second method is pulse oximetry, which assesses blood oxygen saturation by measuring hemoglobin levels.

The artificial intelligence system was able to identify all clinical cases of moderate or severe sleep apnea. In comparison, the standardized questionnaire missed more than 15% of severe breathing disorder cases during sleep. Pulse oximetry identified all severe cases but failed to recognize moderate severity cases.

The new technology, tentatively named OxyDOSA, is already preparing for clinical trials. "The method we developed is a reliable and effective way to detect sleep apnea syndrome in large groups of people," says Behar. "In the future, when we develop the corresponding mobile application, this technology will be available to anyone with smartwatches or bracelets equipped with a built-in oximeter, allowing them to conduct accurate self-examinations."