3d Lung Vessel Segmentation In Computed Tomography Angiography Images

In this paper, a novel lung vessel segmentation method is introduced. In this method, some Reference Points (RPs) were determined by making use of the properties of unchangeable anatomical structure. Due to these RPs, truncus, left-right pulmonary artery, lobar segment vessels have been segmented and subsegment vessels have been detected by looking at the differences of intensities in lung region. If there is pulmonary emboli (PE), heart disease, or abnormal tissues, vessel structure doesn’t regularly continue and decreases the sensitivity of segmentation. Using RPs, vessel structure becomes more definite and sensitivity of the segmentation increases. CTA images belonging 30 patients including different disease are examined and 95% of sensitivity is obtained. The performance of the method for lung vessel segmentation is found to be quite well for radiologists and it gives enough results to the surgeries medically.