Airway identification within planar gamma camera images using computer models of lung morphology.

PURPOSE: Quantification of inhaled aerosols by planar gamma scintigraphy could be improved if a more comprehensive assessment of aerosol distribution patterns among lung airways were obtained. The analysis of planar scans can be quite subjective because of overlaying of small, peripheral airways with large, conducting airways. Herein, a computer modeling technique of the three-dimensional (3-D) branching structure of human lung airways was applied to assist in the interpretation of planar gamma camera images. METHODS: Airway dimensions were derived from morphometric data, and lung boundaries were formulated from scintigraphy protocols. Central, intermediate, and peripheral regions were superimposed on a planar view of the 3-D simulations, and airways were then tabulated by type, number, surface area, and volume in each respective region. RESULTS: These findings indicate that the central region, for example, consists mostly of alveolated airways. Specifically, it was found that alveolated airways comprise over 99% of the total number of airways, over 95% of the total airway surface area, and approximately 80% of the total airway volume in the central region. CONCLUSIONS: The computer simulations are designed to serve as templates that can assist in the interpretation of aerosol deposition data from scintigraphy images.

Three-dimensional simulations of airways within human lungs.

Information regarding the deposition patterns of inhaled particles has important implications to the fields of medicine and risk assessment. The former concerns the targeted delivery of inhaled pharmacological drugs (aerosol therapy); the latter concerns the risk assessment of inhaled air pollutants (inhalation toxicology). It is well documented in the literature that the behavior and fate of inhaled particles may be formulated using three families of variables: respiratory system morphologies, aerosol characteristics, and ventilatory parameters. It is straightforward to propose that the seminal role is played by morphology per se because the structures of individual airways and their spatial orientations within lungs affect the motion of air and the trajectories of transported particles. In previous efforts, we have developed original algorithms to describe airway networks within lungs and employed them as templates to interpret the results of single photon emission computed tomography (SPECTs) studies. In this work, we have advanced the process of mathematical modeling and computer simulations to produce three-dimensional (3D) images. We have tested the new in silico model by studying two different branching concepts: an inclusive (all airways present) system and a single "typical" pathway system. When viewed with the glasses supplied with this volume, the 3D nature of airway branching networks within lungs as displayed via our original computer graphics software is clear. We submit that the new technology will have numerous and seminal functions in future medical and toxicological regimens, the most fundamental being the creation of a platform to view natural 3D structures in vivo with related biological processes (e.g., disposition of inhaled pharmaceuticals).

In silico modeling of asthma.

The incidence of asthma is increasing throughout the world, especially among children, to the extent that it has become a medical issue of serious global concern. Appropriately, numerous pharmacologic drugs and clinical protocols for the treatment and prophylaxis of the disease have been reported. From a scientific perspective, a review of the literature suggests that the targeted delivery of an aerosol would, in a real sense, enhance the efficacy of an inhaled medicine. Therefore, in accordance with published data we have developed a mathematical description of disease-induced effects of disease on airway morphology. A morphological algorithm defining the heterogeneity of asthma has been integrated with a computer code that formulates the behavior and fate of inhaled drugs. In this work, predicted drug particle deposition patterns have been compared with SPECT images from experiments with healthy human subjects (controls) and asthmatic patients. The asthma drug delivery model simulations agree with observations from human testing. The results indicate that in silico modeling provides a technical foundation for addressing effects of disease on the administration of aerosolized drugs, and suggest that modeling should be used in a complementary manner with future inhalation therapy protocols.

A computer model of lung morphology to analyze SPECT images.

Measurement of the spatial distribution of aerosol deposition in human lungs can be performed using single photon emission computed tomography (SPECT). To relate deposition patterns to real lung structures, a computer model of the airway network has been developed. Computer simulations are presented that are compatible with the analysis of SPECT images. Computational techniques that are consistent with clinical procedures are used to analyze airways by type and number within transverse slices of the lung volume. The computer models serve as customized templates, which when analyzed alongside gamma scintigraphy images, can assist in the interpretation of human test data.