Clinical Evaluation of an Automatic Method for Segmentation and Characterization of the Thoracic Aorta with and Without Aneurysm Patients

摘要

The aim of this study was to clinically evaluate a fully automatic method for segmentation and characterizing the thoracic aorta. Prom 2010 to 2013, a total of 27 patients were randomly selected for the study. The automatic method was compared with two segmentations manually performed by two independent radiologists amd a commercial software for gauging calibres. The results of the segmentation measurements showed a Dice Similarity Coefficient of 93.98 ± 0.27, a correlation coefficient of 0.875 ± 0.012, and an intraclass correlation coefficient of 0.909 ± 0.011 for both experts. For the diameter measurements, the values were a Pearson's Correlation Coefficient of 0.933, and an intraclass correlation of 0.913. The Bland-Altman plots showed no statistically significant differences between the commercial method and the proposed method. In conclusion, the method developed is capable of automatically calculating the diameters, and the segmentation of the thoracic aorta.