Eur. Phys. J. Plus (2021) 136: 844
https://doi.org/10.1140/epjp/s13360-021-01764-7

Regular Article

Development of anatomically based structure for human acinus by Lindenmayer system: accurate model for gas exchange in human lung

Department of Chemical and Petroleum Engineering, Sharif University of Technology, P. O. Box 11365-9465, Tehran, Iran

^b rbozorgmehry@sharif.edu

Received: 9 May 2021
Accepted: 16 July 2021
Published online: 15 August 2021

Abstract

Acinar region consists of ~ 33 million airways and provides low resistance against the gas exchange with blood. A three-dimensional anatomically accurate model of pulmonary acinus is useful for simulation of fluid distribution, gas exchange, particle deposition, drug delivery, and detection of structural abnormalities in the lungs. In this study, the stochastic parametric Lindenmayer system has been used to generate the respiratory airways filling a given space. This model takes into account the mechanical properties and details of human acinus which can accurately predict the gas distribution throughout the lungs for the first time. The procedure of finding the dimensions and orientations of airways depends on several decision variables which have been found by particle swarm optimization algorithm. A novel procedure to generate the distribution of alveoli over airways has also been proposed. The morphometric characteristics (i.e., length, diameter, and orientation of airways) of the generated structure are successfully validated with their measured counterparts published in the literature. To examine the degree of asymmetry in the developed structure, it has been used for the prediction of the inert gas washout curve (i.e., the curve of exhaled concentration of inert gas against exhaled volume). This is due to the fact that the slope at the end part of this curve (i.e., alveolar slope) and its progression through the breaths is highly function of the asymmetric pattern of the human bronchial tree. The comparison of predicted slopes with experimental data indicates that the proposed method outperforms the previously reported models. For instance, the relative error of previous models in the prediction of the first slope is ~ 87%, whereas this study gives accurate results.