RESUMO
For optimum therapeutic response from drug administered to the lungs, it is paramount that the aerosolised drug is able to deposit in the lower airways. The filtering characteristics of the respiratory tract, however, make this a particularly challenging task. Computational tools afford a cost-effective means of studying the problem, and here we report on the development of a rapid and reliable method for predicting the pattern of deposition of polydisperse aerosols within human lungs using artificial neural networks (ANNs). Literature (experimental) data on lung deposition of monodisperse aerosols were used to train a single ANN to allow for simultaneous predictions of regional and total aerosol particle deposition patterns in human lungs. When used in modelling the fate of polydisperse aerosols in human lungs, the trained ANN was found to give highly accurate predictions for all lung regions, and all (pharmaceutically relevant) particle sizes and breathing conditions (with errors typically <0.025%). Further testing of the ANN, using 'unseen' in vitro and in vivo data, gave good agreement of lung dosages. It is thus concluded that the ANN produced can be used to provide highly reliable estimates of particle deposition from polydisperse pharmaceutical aerosols generated from breath-actuated dry powder inhalers, nebulizers and metered dose inhalers with spacers.
Assuntos
Aerossóis/farmacocinética , Pulmão/metabolismo , Redes Neurais de Computação , Simulação por Computador , Humanos , Espaçadores de Inalação , Inaladores Dosimetrados , Modelos Biológicos , Tamanho da Partícula , Reprodutibilidade dos TestesRESUMO
PURPOSE: To develop a rapid and reliable method for predicting the pattern of aerosol particle deposition within the human lungs, using artificial neural networks (ANNs). METHODS: Experimental data from the literature were used to train multi-layer perceptron (MLP) networks to allow for prediction of regional and total aerosol particle deposition patterns in human lungs. These data covered particle sizes in the range 0.05-15 microm and three different breathing patterns (ranging from "quiet" breathing to breathing "under physical work conditions"). Three different MLPs were trained, to provide separate predictions of aerosol particle deposition in the laryngeal, bronchial, and alveolar regions. The total deposition fraction for a given set of breathing conditions was computed simply as the sum of the outputs produced from the corresponding regional deposition MLPs. RESULTS: The ANNs developed are shown to give highly accurate predictions for both regional and total aerosol deposition patterns for all particle sizes and breathing conditions (with errors typically less than 0.04%). CONCLUSIONS: We conclude that the current set of ANNs can be used to give good predictions of particle deposition from polydisperse pharmaceutical aerosols generated from breath-actuated dry powder inhalers, nebulizers, and metered dose inhalers with spacers.