QUANTITATIVE-ANALYSIS OF COMPUTED-TOMOGRAPHY SCANS OF THE LUNGS FOR THE DIAGNOSIS OF PULMONARY-EMPHYSEMA - VALIDATION-STUDY OF A SEMIAUTOMATED CONTOUR-DETECTION TECHNIQUE

RATIONALE AND OBJECTIVES. TO develop an analytic software package base d on automated contour detection for the objective and reproducible as sessment of emphysema from computed tomography (CT) scans. METHODS. A semiautomated technique was developed for the definition of lung conto urs in CT cross-sections followed by the assessment of pulmonary CT pa rameters describing the disease state, For 78 images, the semiautomate d contour detection was performed and compared with contours drawn by an experienced radiologist by calculating the systematic area differen ce (bias) and differences in pulmonary CT parameters such as the mean lung density (MLD). In addition, intraobserver and interobserver varia bilities were determined in a subset of 15 images. RESULTS. The areas enclosed by the semiautomatically detected contours were slightly larg er than the manual ones (bias < 2.1%). The biases in the observer stud ies were smaller in the semiautomated versus the manual case (0.3% vs. 1.3%). The standard deviation of the MLD differences with a manual an alysis was larger by a factor of five than in the semiautomated case, On average, manual analysis required 2 minutes, 18 seconds per lung; t his time was reduced to 11.5 to 29 seconds with the semiautomated appr oach, depending on the respiration state. CONCLUSIONS. The semiautomat ed approach is preferred over the manual approach because of its highe r consistency and its shorter analysis time.