Improving the segmentation of lung vessel trees by reducing the uncertainty in neutrosophic transform

Most applications in the field of medical image processing require precise estimation. Efficient and automatic image segmentation methods are useful for the isolation and visualization of vessels in computed tomographic angiography (CTA). There have been many methods proposed for the segmentation of vessels. To achieve this goal, this work aims to improve the segmentation of lung vessel trees by reducing the uncertainty in 3D Frangi filter and 3D neutrosophic transform. First of all, gray-level thresholding and some morphological processes are applied to have the segmentation of lung region mainly. Next, 3D Frangi filter is applied to detect lung vessel trees. Finally, 3D neutrosophic transform integrated with k-means clustering obtains better detection of lung vessel trees.