Pulmonary nodule automatic detection method and system based on chest CT image

Abstract

The invention discloses a pulmonary nodule automatic detection method and system based on a chest CT image; the method comprises the following steps: receiving a chest CT image, using a preset 3D convolution nerve network to carry out pulmonary nodule focus detection for the chest CT image, and outputting a detection result; receiving the CT image and the detection result, and loading the detection result with the corresponding chest CT image to form a fused file; obtaining the fused file and carrying out visualization treatment, providing browse auxiliary operations for users on the fused file with the visualization treatment, and providing a calculation and measuring function. The method and system can isolate lung areas, extract candidate pulmonary nodules and remove false positives, thus improving focus detection efficiency, and the false positive efficiency can reach 97.7%. the method can use visualization three dimensional forms to directly assist a doctor to determine the pulmonary nodule forms.

Description

Technical field [0001] The invention relates to the technical field of medical imaging computer-assisted diagnosis, in particular to a method and system for automatically detecting lung nodules based on chest CT images. Background technique [0002] The automatic detection technology of lung nodules in chest CT images is designed to automatically find nodules in the images while minimizing false positives. The existing lung nodule detection technology is generally divided into four steps: image preprocessing, lung region segmentation, candidate nodule extraction and false positive elimination. [0003] Chest CT images from different sources are usually different in layer spacing and pixel scale, so they need to be preprocessed to unify the scale. Lung region segmentation can narrow the detection range, which not only reduces time consumption, but also effectively reduces false positives. Commonly used lung region segmentation algorithms include gray threshold segmentation, connec...

AI Technical Summary

Problems solved by technology

There are three main problems in this type of algorithm: 1) The sensitivity of ground glass nodules is low. Compared with solid nodules, the density of ground glass nodules is lower, and the edges are difficult to extract, which is not only difficult to detect through effective segmentation algorithms. candidate, and lacks an effective feature description for classification

However, there are still two problems with this type of technology: 1) The detection algorithm is not end-to-end, and some artificially designed steps still exist, which leads to certain limitations in the scope of the algorithm; 2) Using 2D projection images as input cannot be effective Leverage contextual information in CT images

Images