A Pulmonary Nodule Auxiliary Detection System and Method Based on Improved Convex Hull Method

Abstract

The invention provides a pulmonary nodule auxiliary detection system and method based on an improved convex hull method, and relates to the field of computer-aided detection. The system includes lung contour segmentation, lung parenchyma cavity filling and tracheal removal, lung boundary repair, suspected candidate nodule segmentation, gray-level co-occurrence matrix construction, Haralick feature parameter calculation, Haralick feature set construction and pulmonary nodule auxiliary detection 8 units; The method includes: acquiring a pulmonary nodule image and preprocessing it; performing binary segmentation on the image; Filter conditions; generate the gray-level co-occurrence matrix and Haralick feature set of eligible images; obtain the trained ELM diagnostic device; obtain the risk probability of the image of the pulmonary nodule to be diagnosed. This method can effectively improve the performance of predicting the risk of pulmonary nodules, assist clinicians to better diagnose patients according to the risk probability, and improve the accuracy of diagnosis.

Description

technical field [0001] The invention relates to the technical field of computer-aided detection, in particular to a pulmonary nodule-assisted detection system and method based on an improved convex hull method. Background technique [0002] At present, lung cancer has become the malignant tumor with the highest mortality rate, mainly because it is difficult to detect in the early stage and difficult to cure in the late stage. The early manifestation of lung cancer is pulmonary nodules. Early detection, early diagnosis and early treatment of pulmonary nodules are of great significance. Clinically, the most commonly used detection method of pulmonary nodules is the imaging detection method of computerized tomography (CT). However, with the development of CT technology, more and more pulmonary nodules can be detected, and it is very difficult to find and diagnose pulmonary nodules only by the visual observation of radiologists. Therefore, the risk prediction system for pulmon...

AI Technical Summary

Problems solved by technology

Lung segmentation methods mainly include segmentation based on threshold method, segmentation based on region growing method and level set method, and some segmentation based on hybrid methods, but these methods do not effectively deal with the convex shape of the inner and outer contours of the lungs.

Images