Overview

• The study published in Current Biology used a wearable ‘Nasal Holter’ device to record nasal airflow over 24 hours in 100 participants and applied machine learning to analyze 24 breathing parameters.
• Researchers achieved 96.8% accuracy in identifying individuals from their breathing patterns during waking hours, with over 95% accuracy maintained up to two years later.
• Analysis of breathing patterns revealed correlations with physical and mental health metrics, including body mass index and self-reported anxiety and depression scores.
• The researchers are working on creating a more comfortable and discreet version of the device to enable widespread health monitoring and potential disease diagnosis.
• The findings highlight breathing as a stable biometric identifier and suggest respiratory fingerprints could offer noninvasive insights into neurological and psychological conditions.