Pulmonary nodule detection method and device, model training method and device, equipment and medium

Abstract

The invention discloses a pulmonary nodule detection method and device, a model training method and device, equipment and a medium. The method comprises the steps that multiple lung CT images to be detected are acquired, geometrical characteristics are extracted from the multiple lung CT images respectively, the geometrical characteristics are used for representing geometrical information of potential nodules, a similarity matrix between the lung CT images is calculated based on the geometrical characteristics, and for each geometrical characteristic, the similarity matrix and the geometric characteristics are input into a preset graph convolutional neural network for processing to obtain image characteristics; and the probability of existence of pulmonary nodules in each lung CT image isdetermined based on the image characteristics. The similarity matrix between the lung CT images is calculated through the geometrical characteristics of the lung CT images, and the potential geometrical information and the similarity relationship between the lung CT images to be detected are fully mined by using the graph convolutional neural network, so that the influence of blood vessels and other lung shadow structures in the lung CT images on the detection result can be reduced, and the detection precision is improved.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of image processing, and in particular to a pulmonary nodule detection method, a model training method, a device, a device, and a medium. Background technique [0002] Affected by factors such as environment, smoking and genetics, lung cancer is the malignant tumor with the highest mortality and morbidity in my country. According to medical data, the 5-year survival rate of early lung cancer is significantly higher than that of advanced lung cancer. Early detection, early diagnosis and treatment are important ways to improve lung cancer. [0003] Judging whether there are pulmonary nodules in the lungs is a powerful indicator for judging cancer, so early screening of pulmonary nodules becomes particularly important. Among them, low-dose chest CT images have the characteristics of thin layers, clear field of vision, and few interference factors. Therefore, the detection of pulmonary...

AI Technical Summary

Problems solved by technology

[0004] However, through the automatic detection of the computer aided diagnosis system, there will be phenomena such as missed detection and false detection.

This is because pulmonary nodules and blood vessels and other lung shadow structures are similar in shape in CT images, which will lead to a large number of false positives and affect the detection results

Images