Investigating frequency contents of capnogram using fast fourier transform (FFT) and autoregressive modelling

Abstract

In this study, the frequency contents of capnogram were investigated. Capnogram is the graphical output of capnography that represents the different changes in expiratory volume. Capnography is generally used for the monitoring of carbon dioxide (CO2) level during respiration. This method is not only simple, non- invasive and relatively inexpensive, but also mandated or recommended for patient monitoring during clinical procedures by medical societies representing anaesthesiology, cardiology, critical care, paediatrics, respiratory care and emergency medicine. Hence, the signal processing and analysis of capnogram will help in understanding its nature for the diagnosis and prognosis of a variety of respiratory disorders. It should be noted that till now there is no attempt to investigate the frequency contents of capnogram. Therefore, we investigated the frequency properties of capnogram to lead towards better and more accurate diagnostic algorithms related to respiratory malady. Fast Fourier transform (FFT) and autoregressive (AR) modelling-Burg method were used to calculate the power spectral density (PSD) in both normal and asthmatic capnograms, and the preliminary results showed that the frequency properties of the capnograms were significant to distinguish between asthmatic and non-asthmatic patients. These results revealed the potential of using the frequency contents of capnogram as a diagnostic tool or indicator, thus significantly helping medical practitioners involved in respiratory care.