A detailed comparison of particle deposition of a female smoker lung compared to a nonsmoker is vital when trying to gain a further understanding of lung cancer. To make this comparison, two airway models were created that consisted of the oral cavity, oropharynx, laryngopharynx, larynx, trachea, and generations up to the 3rd, 4th or 5th generation. The difference between the two models is in the shape of the oral cavity. The smoker model has an oral cavity that represents the mouth during the inhalation of a cigarette and the nonsmoker model represents normal oral breathing. The oral cavity models were created by scanning a cast. The model of the oropharynx, laryngopharynx, and larynx was created by a medical illustrator based on dimensions from medical photographs and a cast of this region. The trachea through the 5th generation was created by segmenting the 2D cryosectioned images from the Visible Female Dataset and reconstructing them to create a 3D model. It was determined that the error of the segmentation process was 0.33 mm based on the pixel size of the images. The model was then smoothed, decimated and trimmed, which incurred no error beyond the segmentation error. The airway created was compared to values from the literature and it was found that overall, the trachea, main bronchi and lobar bronchi, where the data was recorded as asymmetric, were in best agreement with the asymmetric morphology given by Horsfield et al. (1971). The 2nd through the 5th generation, where the dimensions were recorded as symmetric dimensions, were most closely related to the symmetric data given by Weibel (1964) and Phalen et al. (1985). It was determined that this model was a good representation of the population and a Computational Fluid Dynamic study was preformed. The mesh of the smoker model consisted of 1.42 million cells, where as, the nonsmoker model had 1.41 million cells. The surface grid was examined by checking the y+ values, and grid convergence was determined. The turbulent SST k-co solver was used with second order accuracy. The two models were solved with the same parameters and boundary conditions with the exception of the inlet velocity magnitude. The smoker model had a velocity magnitude of 9.468 m/s and the nonsmoker had a velocity magnitude of 4.962. Velocity contours and turbulence was higher in the smoker model than in the nonsmoker model. Total, regional and local deposition was determined for both models by injecting 50,000 inert unit density spheres at the inlet. 45 % of particles deposited in the smoker model, where as, 21 % deposited in the nonsmoker model. Regional deposition showed that particles were concentrated in the back of the throat, larynx and in the bifurcation regions. The high percentage of deposition in the smoker model could explain the amount of oral cavity, larynx, lung and bronchi cancer. Also, the limited amount of deposition in the nonsmoker model shows that more particles travel deeper into the lungs where more gas exchange occurs and translocation of particles to other organs seems more likely in this case.
Library of Congress Subject Headings
Tobacco--Physiological effect--Simulation methods; Women--Tobacco use--Simulation methods; Lungs--Anatomy--Data processing; Lungs--Models; Atmospheric deposition--Physiological effect--Simulation methods; Particle image velocimetry
Department, Program, or Center
Manufacturing and Mechanical Engineering Technology (CAST)
Russo, Jackie, "3d reconstruction of a female lung using the Visible Human Data Set to predict cigarette smoke particle deposition" (2008). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus