Jin Fei

Human respiratory function studies, namely the breathing rate and volume modeling, can indicate the overall health status of a person. The Mid-Wave Infra-Red (MWIR) imaging system can sense the breathing thermal signal based on the radiation information within nostril and nasal vestibule regions. The breath signal is quasi-periodic due to the interleaving of high and low intensities corresponding to expirations and inspirations respectively.

We have developed various non-contact methods to measure human breathing rate from different aspects. We applied resizable window on re-sampled breathing thermal signal to determine the breathing frequency through Fourier analysis in the carbon dioxide absorption band. In addition, we compute the mean breathing cycle and use the harmonic components to characterize the breathing pattern. The harmonic analysis highlights the intra-individual similarity of breathing patterns.

We have performed experiments on subjects at distance ranging from 6-8 ft. We compared the experimental results computed by our novel method with ground-truth measurements obtained via a traditional contact device (PowerLab/4SP from ADInstruments with an abdominal transducer).

Novel non-contact passive thermal imagery to sense breathing opens the way for desktop, unobtrusive monitoring of an important vital sign, that is, breathing rate. It may find widespread applications in preventive medicine as well as sustained physiological monitoring of subjects suffering from chronic respiratory ailments, like apnea. The thermal imagery has been used in mask leakage detection.