Intelligent image processing method for lung shadow

Abstract

The invention provides an intelligent image processing method for a lung shadow, and the method comprises the steps: S1, extracting an original image of a lung, carrying out the preprocessing of the image, and segmenting the image to obtain a lung nodule; S2, performing feature extraction on the lung nodules to obtain feature parameters; S3, performing feature selection on the lung nodules by adopting a principal component analysis strategy; and S4, identifying the lung nodules by adopting a classification algorithm to obtain a positive data set and a negative data set. According to the method, the region structure where the lung nodule is adhered to the region outside the lung in image display is captured and supplemented through the edge filling algorithm, and the reliability of data during lung image processing is improved.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an intelligent image processing method for lung shadows. Background technique [0002] At present, lung cancer accounts for a high mortality rate of cancer diseases. According to statistics in 2015, the total number of newly diagnosed lung cancer cases in China is about 787,000, and about 630,500 lung cancer patients died, which is equivalent to an average of more than 1,700 deaths per day. Lung cancer The total number of patients ranks first in the world and the incidence of lung cancer is increasing day by day, seriously endangering the health of Chinese people. In recent years, the development of technologies such as high-resolution computed tomography, PET-CT, bronchoscopy, and percutaneous tumor puncture has improved the accuracy of precise screening and staging of lung cancer. PET-CT is of high value not only for overall assessment, especially for possible distant...

AI Technical Summary

Problems solved by technology

[0003] Pulmonary nodules, as the basic representation of lung cancer lesions, are an important basis for lung cancer screening. With the continuous improvement of CT image scanning technology, it is urgent to solve the problem of matching image recognition processing technology with it. In terms of extracting the integrity of lung features It is particularly critical, because the high-density lung edge adhesion affects the edge extraction of the lung image, and the similarity of the lung edge tissue is high. The traditional edge extraction algorithm will miss some features, resulting in The edges of the lungs with characteristic parameters are excluded, resulting in data loss during image processing

Images