Scientists have revealed a new biometric marker: your nasal breathing pattern. This is uniquely yours. According to new research, nasal airflow can identify individuals with 96.8% accuracy. That is on par with voice recognition.
The discovery came from a team at the Weizmann Institute of Science. They created a wearable device that tracks nasal breathing patterns over 24 hours. Unlike short clinical tests, this method captures long-term characteristics.
“You would think that breathing has been measured and analyzed in every way,” said Noam Sobel. “Yet we stumbled upon a completely new way to look at respiration. We consider this as a brain readout.”
The study involved 97 participants and used 24 specific features extracted from their breathing patterns. Each feature remained stable over months, even up to two years. This suggests it comes from the brain’s respiratory network.
Participants wore a silicone-encased device on their neck all day. It connected to a nasal cannula with separate tubes for each nostril. This setup recorded every breath in real time, logging airflow asymmetries and temporal pauses.
Researchers used AI models to analyze features like inhale volume and duty cycles. With just one hour of breathing data, they could still identify about 43% of participants during sleep. When they tested data across months, accuracy remained high: 95% during waking hours and over 70% during sleep.
Importantly, they ruled out motion as a confounding factor. Position sensors built into the device could only identify people with 37% accuracy. The breathing pattern did far better.
The study didn’t just identify people. It predicted health and emotional traits. Breathing patterns reflected body mass index, sleep quality, and mental health. For example, those with higher anxiety scores had shorter inhales and more varied pauses during sleep.
These effects were not linked to illness. All participants were healthy and had no clinical diagnoses. Yet their breathing revealed differences tied to depression, anxiety, and even traits linked to autism.
“Perhaps the way you breathe makes you anxious or depressed,” said Sobel. “If that’s true, we might be able to change the way you breathe to change those conditions.”
Researchers used a technique called canonical correlation analysis. It showed a strong link between breathing patterns and psychological traits, with a correlation of 0.87 across multiple tests.
Using nasal airflow data, researchers could predict BMI with strong accuracy. Tidal volume and exhale volume during sleep both correlated with higher BMI. Interestingly, even nasal cycle features, like airflow differences between nostrils, predicted BMI. That suggests brain-driven breathing control is involved.
In those with higher depression scores, the study found faster airflow and longer exhale pauses during wakefulness. For anxiety, shorter inhales and less stable breathing occurred during sleep. Even those with higher autism trait scores showed longer inhale pauses and more variable patterns during sleep.
These findings suggest that breathing patterns reflect both state (like stress or sleep) and trait (like body composition or emotional tendencies).
Despite the promise, the device has limitations. The nasal cannula, often associated with illness, may discourage everyday use. It also doesn’t track mouth breathing and sometimes slips during sleep. Soroka and Sobel plan to improve the device’s design.
Still, the potential is big. A nasal breathing pattern reader could someday be used to track well-being, monitor mental health, or guide therapies. “We definitely want to go beyond diagnostics to treatment, and we are cautiously optimistic,” stated Sobel.
Even now, the researchers can identify you by your breathing pattern. Soon, they might help you breathe your way to better health.
The study is published in the journal Current Biology.
—–
Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates.
Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.
—–