Computer device and method for realizing pulmonary nodule image classification detection

Abstract

The invention provides a computer device and method for realizing pulmonary nodule image classification detection, and the method comprises the steps: scanning a lung CT image of a patient through a CT scanner, and carrying out the adaptive form segmentation of the lung CT image to obtain a pulmonary nodule image; marking the pulmonary nodule images into different categories, and storing the marked pulmonary nodule images into an image database; establishing a pulmonary nodule image unit library based on the pulmonary nodule images in the image database; calculating the distance between everytwo image units in the pulmonary nodule image unit library to obtain a distance matrix; clustering the distance matrix; calculating pulmonary nodule CT value density distribution characteristics of each pulmonary nodule image unit; based on a supervised machine learning model, using CT value density distribution characteristics to realize training and classification of pulmonary nodule risk degrees; and eliminating false positive pulmonary nodule images according to the CT value density distribution of each pulmonary nodule image. According to the invention, the accuracy of pulmonary nodule image classification detection is improved, and the method has a wide application range.

Description

technical field [0001] The invention relates to the technical field of pulmonary nodule image processing, in particular to a computer device and method for realizing classification and detection of pulmonary nodule images. Background technique [0002] In recent years, lung CT images have become more and more widely used in the clinical diagnosis of pulmonary nodules. It is of great significance to analyze the risk of pulmonary nodules through lung CT images, especially for the study of early pulmonary nodules. A high-precision, clinically meaningful, and robust pulmonary nodule image classification and detection system is becoming more and more important. Due to the complexity of pulmonary nodules, the current pulmonary nodule classification and detection technology based on lung CT image processing is not complete. The existing technology is not practical for the classification of pulmonary nodules, which affects the low accuracy of classification and detection of pulmona...

AI Technical Summary

Problems solved by technology

[0003] The main purpose of the present invention is to provide a computer device and method for realizing the classification and detection of pulmonary nodules images, aiming to solve the problem that the lack of completeness of the existing lung nodule classification and detection technology based on lung CT image processing affects the classification of pulmonary nodules. The problem of low detection accuracy

Images