CAD system for quantitative evaluation of chronic obstructive pulmonary disease based on 3-D CT images

Abstract

This paper presents a CAD system for quantitative evaluation of chronic obstructive pulmonary disease (COPD) based on 3-D chest X-ray CT images. In the system, distribution pattern of COPD is automatically analyzed and its result is presented to medical doctors. The system analyzes COPD by using true 3-D image processing techniques based on precise 3-D chest X-ray CT images. In the proposed CAD system, low attenuation areas (LAAs), which are clearly observed in the case of COPD, are automatically extracted by intensity threshold and intensity variance. The extracted regions are analyzed by using 3-D distance transformation. The analyzed results are also visualized by volume rendering and digital radiography reconstruction. The experimental results show that the proposed system is very useful of quantification of COPD diagnosis.