Introduction

Abstract

The analysis of pulmonary images is of eminent importance in modern medicine. Computed tomography (CT) scans of the thorax, for example, enable a three-dimensional insight into the patient’s anatomy and are used to non-invasively study morphology and function of the lung with application to diagnosis, treatment and monitoring of various pulmonary diseases. In this context, the role of techniques for computeraided image analysis is manyfold. For example, they are used to extract quantitative values about the type and growth of pulmonary nodules to early diagnose malignant neoplasms of lung and bronchus [Armato and Sensakovic 2004], which are responsible for the most cancer-related deaths in the US [ACS 2011]. After diagnosis, lung CT scans are used to plan and guide interventions or radiotherapy and to monitor success of the treatment [Yan et al. 1999]. Functional pulmonary image analysis is further used for a regional assessment of ventilation and perfusion parameters to quantify gas exchange [Hoffman et al. 2004], which is of interest for the diagnosis of fibrosis or Chronic Obstructive Pulmonary Diseases (COPD) such as emphysema [Scatarige et al. 2003].