A Recurrent Parameter Model to Characterize the High-Frequency Range of Respiratory Impedance in Healthy Subjects

• Published in: IEEE transactions on biomedical circuits and systems Vol. 7; no. 6; pp. 882 - 892
• Main Authors: Ionescu, Clara M., Hernandez, Andres, De Keyser, Robin
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.12.2013; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Airway Resistance - physiology; Algorithms; Biomedical measurement; Circuits; Equipment Design; Forced oscillations; fractal structure; Fractals; fractional order Laplace; frequency response; High frequencies; Humans; Impedance; ladder network; Ladders; Laplace equations; Lungs; Mathematical models; Networks; Oscillations; Oscillators; Recurrent neural networks; Respiratory Function Tests - instrumentation; Respiratory Function Tests - methods; Respiratory impedance; Respiratory Mechanics - physiology; Respiratory system; upper airway shunt
• ISSN: 1932-4545; 1940-9990; 1940-9990
• DOI: 10.1109/TBCAS.2013.2243837

In this work, a re-visited model of the respiratory system is proposed. Identification of a recurrent electrical ladder network model of the lungs, which incorporates their specific morphology and anatomical structure, is performed on 31 healthy subjects. The data for identification has been gathered using the forced oscillation lung function test, which delivers a non-parametric model of the impedance. On the measured frequency response, the ladder network parameters have been identified and a fractional order has been calculated from the recurrent ratios of the respiratory mechanics (resistance and compliance). The paper includes also a comparison of our recurrent parameter model with another parametric model for high frequency range. The results suggest that the two models can equally well characterize the respiratory impedance over a long range of frequencies. Additionally, we have shown that the fractional order resulting from the recurrent properties of resistance and compliance in the ladder network model is independent of frequency and is not biased by the nose clip wore by the patients during measurements. An illustrative example shows that our re-visited model is sensitive to changes in respiratory mechanics and the fractional order value is a reliable parameter to capture these changes.