Towards a multi-scale computer modeling workflow for simulation of pulmonary ventilation in advanced COVID-19

Physics-based in silico modelling of microvascular pulmonary perfusion in COVID-19

“…Here, we present a model of pulmonary perfusion coupled to a previously described model of pulmonary ventilation. 20 Acinar perfusion was incorporated in the C + + simulation package CHASTE (Cardiac, Heart, and Soft Tissue Environment). 21,22 The development of the patient-specific physics-based pulmonary ventilation model including imaging, segmentation of the CT image, geometry development, airflow and COVID-19 acinar tissue mechanics are outlined in a previous paper by Middleton et al 20 and the ''Diseased 20-60'' model, where ground-glass opacities (GGO) and consolidated regions were assigned 20% and 60% reductions in surfactant respectively is utilized in this study (Figure 1(b)).…”

“…20 Acinar perfusion was incorporated in the C + + simulation package CHASTE (Cardiac, Heart, and Soft Tissue Environment). 21,22 The development of the patient-specific physics-based pulmonary ventilation model including imaging, segmentation of the CT image, geometry development, airflow and COVID-19 acinar tissue mechanics are outlined in a previous paper by Middleton et al 20 and the ''Diseased 20-60'' model, where ground-glass opacities (GGO) and consolidated regions were assigned 20% and 60% reductions in surfactant respectively is utilized in this study (Figure 1(b)). Secondary analysis of the existing and already segmented CT images 20 of a 51-year-old male subject hospitalized with a severe case of COVID-19 were utilized in this study.…”

“…Overall, seven simulations were run to model perfusion in the lung. A hypothetical healthy lung simulation was run along with a disease state with only ventilation defects (using ''diseased 20-60'' case from Middleton et al 20 ). Five additional disease states were run that applied changes in perfusion.…”

“…To this end, the sheet flow model of blood flow in the alveolar capillaries is coupled to a previously described acinar ventilation model. 20 The modelling approach incorporates CT data to determine lung geometry and distribution of heterogeneous damage in acinar tissues (as described in Middleton et al 20 ) as well as previously reported DE CT data 5 on pulmonary perfusion changes to provide a patient-centred approach to studying COVID-19 lung mechanics and perfusion. Herein, we apply this hypothesis-testing platform to investigate the effects of various and conflicting findings on acinar level pulmonary perfusion abnormalities on global lung function.…”

Feature selection of pre-trained shallow CNN using the QLESCA optimizer: COVID-19 detection as a case study

“…In recent years, the application of artificial intelligence in the automatic diagnosis of medical images has yielded promising results. Deep convolutional neural networks have been used in previous studies to classify chest X-ray and CT images and to successfully diagnose common chest diseases [ 1 , 45 – 47 ]. In the current outbreak of COVID-19, researchers are attempting to alleviate the epidemic through their research.…”