Multifractal Behaviour of Respiratory Signals

Abstract

In this study, to analyze the biomedical signals emerging from fractal structures in the human body, fractal analysis was used. Respiratory signals, such as airflow, mouth pressure, and lung volume, comprise a complex relationship that has not been inspected to date. Furthermore, the mechanism for which it is linked to the lung’s fractal structure has not been scrutinized to date. Thus, using a well-known method, known as multifractal detrended fluctuation analysis (MF-DFA), this study aims to determine both mono- and multi-fractal property of respiratory signals ,. The real signals were analyzed using the MF-DFA algorithm. Moreover, for different scales, generalized Hurst exponent values were calculated. The results demonstrated that respiratory signals are fractional Brown motion-type signals, whereas fractal properties demonstrate less intersubject change. Moreover, in addition to both airflow and lung volume, respiratory signals and sounds are multifractal signals. In conclusion, the presence of the lung’s long-memory property is the primary reason of multifractality.