Pulmonary tuberculosis intelligent recognition method and system with image symptom interpretation

Abstract

The embodiment of the invention provides a pulmonary tuberculosis intelligent recognition method and system with image symptom interpretation. The pulmonary tuberculosis intelligent recognition methodcomprises the steps: preprocessing an X ray chest radiograph so as to be converted into a vector diagram; carrying out abnormal region identification to obtain a classification result of whether a suspected focus exists or not; judging whether a suspected focus is identified in the abnormal area or not; performing classification correction processing to obtain a conditional probability of the suspected lesion; judging whether a suspected focus caused by pulmonary tuberculosis exists in the abnormal area or not; processing a suspected area corresponding to the suspected focus on an original image to obtain a sub-image vector diagram; explaining that the abnormal region has image characterization significance to obtain characterization description to judge whether the abnormal region has pulmonary tuberculosis or not; and based on the eigenimage description, carrying out pulmonary tuberculosis activity discrimination. According to the embodiment of the invention, the internal relation between the image symptom features in the chest radiography can be effectively obtained, and the judgment logic of iconography can be better met compared with the judgment made only based on the image,and the recognition precision and efficiency can be greatly improved.

Description

technical field [0001] The invention relates to the technical field of medical image processing, in particular to an intelligent identification method and system for pulmonary tuberculosis with image sign interpretation. Background technique [0002] Tuberculosis, as an infectious disease that poses a great threat to human health, has always attracted people's attention. According to the World Health Organization (WTO), in 2017 alone, tuberculosis (TB) caused about 1.3 million deaths, most of them in developing countries. In view of the raging TB, the WTO proposes a combination of prevention and treatment, one of the important links is to carry out X-ray chest X-ray screening for TB. Although it is of great significance to develop TB chest X-ray screening, it is still difficult to carry out universal TB chest X-ray screening due to the limitation of chest X-ray images and the lack of experienced physicians in developing countries. In underdeveloped areas, TB chest X-ray sc...

AI Technical Summary

Problems solved by technology

[0004] At present, the general TB identification method based on artificial intelligence often uses a patient's chest radiograph to make a judgment, which is often based on the abnormal identification or judgment made by the picture and the model itself, while ignoring the shape of the abnormal area itself clinical implications, and these identification methods do not take into account the intrinsic relationship of lesion evolution contained between imaging features on chest radiographs

The above shortcomings make it difficult for general methods to give acceptable reasons for identification or judgment, and because of ignoring the intrinsic relationship between image representations, these methods are difficult to give acceptable judgments of TB activity

Images