AI Can Now Detect Sleep Apnea Just From a Heart Test

Researchers at Mayo Clinic have developed an artificial intelligence (AI) algorithm capable of detecting obstructive sleep apnea (OSA) using results from a standard electrocardiogram (ECG), which could revolutionize early detection, particularly in women.

• Mayo Clinic’s AI can detect sleep apnea from a 12-lead ECG test.
• The model was trained on 11,299 patient ECGs and sleep evaluations.
• Women showed stronger ECG signatures of OSA than men.

Sleep apnea, a common but often overlooked condition, affects more than 936 million adults worldwide, leading to repeated pauses in breathing during sleep and serious cardiovascular risks. However, the condition frequently goes undiagnosed, particularly in women, due to different symptoms than in men.

Mayo Clinic’s new AI algorithm addresses this diagnostic gap. By analyzing standard ECG test results, a widely available and inexpensive tool, the system can detect patterns in the heart’s electrical activity that correlate with obstructive sleep apnea, per a report from November 6.

According to Virend Somers, M.D., Ph.D., Alice Sheets Marriott Professor of Cardiovascular Medicine and senior author of the study:

“OSA affects the heart to the point where AI algorithms can detect the OSA signature from the ECG, which in essence is a representation of the electrical activity of the heart muscle cells.”

Why Women Stand Out in the Data

As it happens, the research team analyzed ECGs from more than 11,000 patients, over 7,000 of whom had a confirmed diagnosis of OSA. Surprisingly, the algorithm was especially effective at detecting the condition in female patients, even though their apnea cases were milder on average. Per Dr. Somers:

“The most surprising finding was the increased visibility on the ECG of OSA in the females compared to the males, even though the OSA severity was less in the females. […] This is relevant since emerging data consistently suggest that females have a greater relative likelihood of suffering the cardiovascular consequences of OSA, even if their OSA may be considered ‘milder’ by standard diagnostic criteria.”

Meanwhile, the study also suggests that women may experience greater damage to heart muscle cells from OSA, a factor that could further explain the heightened cardiovascular impact seen in female patients.

All things considered, this AI tool could not just help clinicians to diagnose OSA, but also to monitor how treatments for it affect cardiovascular health, offering a non-invasive way to track improvements or residual risks over time. With further validation, it could pave the way for routine OSA screening during heart checkups.

More Must-Reads:

• Can AI-Controlled Satellites Stop Global Warming? Elon Musk Says Yes
• Soccer Coach Used ChatGPT for Match Tactics – And it Helped Her Win
• Google Warns of New Malware Using AI to Evade Detection