Pulmonary membrane adhesion nodule region accurate repairing method for pulmonary CT image threshold segmentation result

Abstract

The invention discloses a pulmonary membrane adhesion nodule region repairing method for a pulmonary CT image threshold segmentation result. The method comprises the steps of carrying out the SIFT feature detection on the lung binary image to obtain the feature points in the image, calculating the nearest neighbor substitution points of the obtained non-boundary feature points, and after the calculated substitution points are used for replacing original non-boundary feature points, obtaining a corresponding boundary feature point set; for each boundary feature point in the boundary feature point set, extracting a support boundary based on a specific rule; segmenting the support boundaries according to the calculated connectivity of the support boundaries in the support boundary image to obtain the mutually independent support boundaries in the image; for each support boundary segment, identifying a to-be-repaired boundary by using the curvature code; using a GVF-based method to drive the smooth profile curve for the identified portion to repair the recessed region and boundary. By applying the method provided by the invention, the omissive pulmonary membrane bonding joint region inthe threshold segmentation result can be accurately repaired, and the pulmonary membrane nodule region is re-incorporated into the pulmonary parenchyma region, so that the accuracy of the pulmonary CT image lesion tissue analysis algorithm is ensured.

Description

technical field [0001] The invention belongs to the technical field of lung CT image analysis and processing, and in particular relates to a method for accurately repairing a pulmonary membrane adhesion nodule region aimed at the threshold segmentation result of a lung CT image. Background technique [0002] In the past two decades, the incidence of lung cancer in my country has increased dramatically. In addition, because lung cancer has a high mortality rate among all cancers, it has become an important disease that endangers the health of Chinese residents. For lung cancer, imaging examination using CT equipment is currently an important diagnostic tool. Therefore, recognition technology for lung CT images is worthy of attention. [0003] With the improvement of computing power of computer hardware and the improvement of relevant algorithm theory, the scheme of using computer hardware and algorithm to assist detection of lesion in lung CT images has become more and more...

AI Technical Summary

Problems solved by technology

When the initial condition is poor, it needs to be processed twice or again, so that the algorithm has the problem of low processing efficiency

[0015] Part of the pulmonary nodule area processing algorithm is not perfect for the position processing scheme of the lesion area, and there is a problem that the accuracy of the lesion position recognition is not high

Images