In asthma and COPD, some airways of the tracheo-bronchial tree can be constricted, from moderate narrowing up to closure. Those pathological patterns affect the lung ventilation distribution. While some imaging techniques enable visualization and quantification of constrictions in proximal generations, no non-invasive technique provides precise insights on what happens in more distal areas. In this work, we propose a process that exploits lung ventilation measures to access positions of airways closures in the tree. This identification approach combines the lung ventilation model in which a tree is strongly coupled to a parenchyma description along with a machine learning approach. Based on synthetic data generated with typical temporal and spatial resolutions as well as reconstruction errors, we obtain very encouraging results with a detection rate higher than 90%.