[CT imaging of chronic interstitial lung diseases: from diagnosis to automated quantification]. / Imagerie des pneumopathies infiltratives diffuses chroniques en tomodensitométrie : du diagnostic à la quantification automatisée.

Computed tomography is important for the diagnosis and follow-up of chronic diffuse interstitial lung diseases. Image quality has improved from each generation of scanner to the next and this continues to allow a better characterization of extent of pathology, or even the nature of the pathological process (potentially reversible inflammatory lesions compared to fibrotic lesions). The diagnostic imaging approach has evolved at the same time as technological developments. We initially thought in terms of the predominant lesions (nodular, alveolar consolidation, ground-glass opacity), and then moved to reasoning based on patterns, which are a combination of several elementary lesions (typically for the diagnosis of idiopathic pulmonary fibrosis). Nowadays, studies are focused on building models characterizing a specific disease and which combine several distinct patterns (typically for ground-glass opacity analysis). CT also allows a quantification of the extent of lung disease, which is linked to the prognosis of the disease and helps to monitor its progression. This quantification is usually based on visual criteria, the principles of which are summarized here. The development of automated quantification software could in the near future, be a support for the radiologist.

Quantification of bronchial dimensions at MDCT using dedicated software.

This study aimed to assess the feasibility of quantification of bronchial dimensions at MDCT using dedicated software (BronCare). We evaluated the reliability of the software to segment the airways and defined criteria ensuring accurate measurements. BronCare was applied on two successive examinations in 10 mild asthmatic patients. Acquisitions were performed at pneumotachographically controlled lung volume (65% TLC), with reconstructions focused on the right lung base. Five validation criteria were imposed: (1) bronchus type: segmental and subsegmental; (2) lumen area (LA)>4 mm2; (3) bronchus length (Lg) > 7 mm; (4) confidence index - giving the percentage of the bronchus not abutted by a vessel - (CI) >55% for validation of wall area (WA) and (5) a minimum of 10 contiguous cross-sectional images fulfilling the criteria. A complete segmentation procedure on both acquisitions made possible an evaluation of LA and WA in 174/223 (78%) and 171/174 (98%) of bronchi, respectively. The validation criteria were met for 56/69 (81%) and for 16/69 (23%) of segmental bronchi and for 73/102 (72%) and 58/102 (57%) of subsegmental bronchi, for LA and WA, respectively. In conclusion, BronCare is reliable to segment the airways in clinical practice. The proposed criteria seem appropriate to select bronchi candidates for measurement.

An image-based computational model of oscillatory flow in the proximal part of tracheobronchial trees.

A computational model of an oscillatory laminar flow of an incompressible Newtonian fluid has been carried out in the proximal part of human tracheobronchial trees, either normal or with a strongly stenosed right main bronchus. After acquisition with a multislice spiral CT, the thoracic images are processed to reconstruct the geometry of the trachea and the first six bronchus generations and to virtually travel inside this duct network. The facetisation associated with the 3D reconstruction of the tracheobronchial tree is improved to get a computation-adapted surface triangulation, which leads to a volumic mesh composed of tetrahedra. The Navier-Stokes equations associated with the classical boundary conditions and different values of the flow dimensionless parameters are solved using the finite element method. The airways are supposed to be rigid during rest breathing. The flow distribution among the set of bronchi is determined during the respiratory cycle. Cycle reproducibility and mesh size effects on the numerical results are examined. Helpful qualitative data are provided rather than accurate quantitative results in the context of multimodelling, from image processing to numerical simulations.