Thomas Laporte - Lung Reconstruction : From CT-Scan to usable mesh of the respiratory system
26 May 2020 –
Tuesday 26 May, 11h(FR)=12h(GR) at https://meet.jit.si/Aromath
Lung Reconstruction : From CT-Scan to usable mesh of the respiratory system
Algorithmic advances, particularly in artificial intelligence and segmentation, now make it possible to develop customized models for the study of the respiratory system.
This work uses the power of deep learning, in order to extract from medical images (CT-Scan) a geometry of the lungs, lobes and, to go further, of the bronchial tree (the first 2-3 generations).
A bronchial tree growth algorithm allows, taking into account the extracted geometries, to obtain the entire bronchial tree. The set of meshes obtained can then be used during simulation to better understand certain mechanical phenomena of the respiratory system.