Automated volume measurements of pulmonary emphysema on 3D chest CT images

Abstract Automatic extraction of the pulmonary emphysema area on 3-D chest CT images was performed based on an adaptive thresholding technique and we proposed a method to estimate the ratio of emphysema area within whole lung volume. We employed 32 cases (15 normal and 17 abnormal) previously diagnosed by radiologists, the ratio of all normal cases were less than 2%, and that of abnormal cases ranged from 1% to 26%. It was concluded that the effectiveness of our approach was confirmed in this initial work.

[1]  Hans Knutsson,et al.  Recognizing emphysema - a neural network approach , 2002, Object recognition supported by user interaction for service robots.