Improvement of fully automated airway segmentation on volumetric computed tomographic images using a 2.5 dimensional convolutional neural net
Multi-detector computerized tomography (MDCT) offers volumetric images of the airway tree geometry at the sublobal level with submillimeter resolution. Quantifying the peripheral geometry from MDCT images is important for diagnosis and treatment planning of pulmonary diseases involving airway pathology, such as chronic obstructive pulmonary disease (COPD), cystic fibrosis, and interstitial lung diseases (Barnes and Hansel, 2004; Pu et al., 2012). Airway tree segmentation plays an especially important role in pulmonary disease analysis because it quantifies the anatomical features, including the airway wall thickening, wall area, lumen area, wall –lumen area ratio, wall–lumen diameter ratio, and changes in lumen diameter.
Source: Medical Image Analysis - Category: Radiology Authors: Jihye Yun, Jinkon Park, Donghoon Yu, Jaeyoun Yi, Minho Lee, Hee Jun Park, June-Goo Lee, Joon Beom Seo, Namkug Kim Source Type: research