Lung image retrieving method and apparatus

Abstract

The invention provides a lung image retrieving method and apparatus. The method comprises the following steps: a type of a lesion contained in a to-be-retrieved lung image is obtained and a corresponding retrieving feature vector is obtained from the to-be-retrieved lung image based on the lesion type; and in a sample image library corresponding to the lesion type, at least one sample lung image with the highest similarity with the to-be-retrieved lung image is obtained based on the retrieving feature vector, wherein the sample image library includes various sample lung images belonging to thesame lesion type and corresponding sample feature vectors obtained from all sample lung images based on the same lesion type. According to the method provided by the invention, corresponding retrieving feature vectors are obtained based on different lesion types in the to-be-retrieved lung image and similar sample lung image retrieving is carried out based on the retrieving feature vector, so that the lung image retrieval accuracy is improved.

Description

technical field [0001] The present invention relates to technical fields such as image retrieval, medical treatment and sanitation, and more specifically, to a lung image retrieval method and device. Background technique [0002] With the continuous development of society and the continuous improvement of living standards, people pay more and more attention to health. How to use relatively scarce medical resources to meet the health needs of the people is an urgent problem to be solved. Lung disease is a disease with a high incidence in the current society. The annual mortality rate of lung cancer is the highest. More and more people pay attention to the screening of lung cancer. However, the lack of professional doctors is an important reason for the difficulty in seeing a doctor. When a patient's lung tissue becomes diseased, doctors need to use professional CT imaging equipment to obtain image information of the patient's lungs and make a diagnosis based on the lung image...

AI Technical Summary

Problems solved by technology

However, lung CT images generally have hundreds of layers, and it takes a lot of energy for doctors to accurately find out the lesions; furthermore, because doctors manually make judgments based on the attributes of lung lesions, the efficiency of browsing images is low, and the level of each doctor exists. There is a gap between high and low, and some doctors need to consult experienced doctors if they cannot judge

[0003] For this reason, in the prior art, in order to improve the accuracy of lung image judgment, doctors can quickly make a diagnostic conclusion by referring to the lung images of similar lesion cases, while in the prior art, doctors manually search for lung images of similar lesion cases. Obtaining the most similar lung images from a large number of historical lung images often has the disadvantages of low efficiency and low accuracy

Images