Extraction method of trachea tree based on trachea segmentation

Abstract

The invention discloses a bronchial tree extraction method based on forward-wave tracheal segmentation, comprising steps of A, obtaining chest CT data, B,creating a bronchial tree structure image and a temporary data pool, C, initializing a preset root node as a first node collection, D,traversing the 26-neighbourhood of the pixel value of each node in a first node collection, E,adding the node where the difference of the CT value and the pixel value is smaller than the first threshold in the 26-neighbourhood into a second node collection, executing the step F when the quantity of the second node collections is zero or the quotient of the quantity of the second node collections and the quantity of the first node collections is greater or equal to a second threshold, F,checking whether the first node collection has a farther node, if not, outputting the pixel values and the CT values of all nodes in the temporary data pool, and obtaining the extracted bronchial tree structure image. The embodiment of the bronchial tree extraction method can accurately and fast extract the Bronchial tree structure, the segmentation result is accurate and the time consumption is less.

Description

technical field [0001] The invention relates to the technical field of computer image processing, in particular to a method for extracting a trachea tree based on trachea segmentation. Background technique [0002] With the development of industrialization in modern society, air pollution has increased, and the incidence of human lungs has become higher and higher, and there are many types, such as lung cancer, bronchitis, emphysema, asthma, etc. With the development of multi-slice spiral CT, radiologists can obtain patient image information through one scan, and analyze and diagnose diseases in combination with image processing technology. CT imaging is considered to be the gold standard for diagnosing lung diseases. However, due to the complex structure of the trachea tree, the contrast is poor, and it is easily affected by noise and volume effects. For example, when the lung trachea is filled with air, the CT value of the human tissue shows that the CT value of the lung t...

AI Technical Summary

Problems solved by technology

At present, the most common trachea segmentation methods use the method of region growth. The threshold selection in this method has a great influence on the segmentation results. When the threshold selection is inappropriate, not only will the segmentation time be too long, but it will easily lead to segmentation failure. The effect is not ideal

Images