Point position association method of ipsilateral aberrant breast molybdenum target images

Abstract

The invention relates to a point position association method of ipsilateral aberrant breast molybdenum target images. The method has the advantages that lesion points in a CC-view compressed image aresubstituted into a mathematical model to obtain the positions of the rotated lesion points on an MLO-view compressed image which is ipsilateral aberrant to the CC-view compressed image, the CC-view compressed image with the lesion points and the MLO-view compressed image are used as the input images to recover a three-dimensional breast image in a free state, the lesion points of the CC-view compressed image are recovered into a lesion curve formed in the free state, MLO angle rotation is performed on the three-dimensional breast image and the lesion curve, MLO-view compression is performed on the rotated three-dimensional breast image and lesion curve, and the projection principle is used to project the lesion curve to the MLO-view compressed image to obtain the positions of the lesion points on the MLO-view compressed image. By the method, correspondence of the same lesion points of the two images is achieved, the point position association relationship of the ipsilateral aberrant breast molybdenum target images is built, and a doctor can be helped to comprehensively analyze the lesion points.

Description

technical field [0001] The present invention relates to the technical field of mammography images, in particular to a point position correlation method of ipsilateral heterotopic mammography images. Background technique [0002] With economic development and improvement of living standards, people pay more and more attention to their own health maintenance and early detection of diseases. As the number one killer of middle-aged and elderly women, breast cancer has an impact on their lives that cannot be ignored. According to statistics, about 1.7 million new breast cancers are diagnosed every year in the world, and the incidence of breast cancer in developed countries is higher than that in developing countries. It is estimated that the number of breast cancer patients will reach about 2.5 million by 2021, and the incidence rate will increase from less than 60 cases / 100,000 women (55-69 years old) to more than 100 cases / 100,000 women (Shen Kunwei, Li Hongwei. Breast Cancer ...

AI Technical Summary

Problems solved by technology

However, the thinking of the above and other classical algorithms is to treat suspected calcification points and masses as a feature object, and the method formed with a similar strategy has at least the following defects: 1) Only compress images for a single CC or MLO position For detection, it is impossible to match the same lesion point in the CC image and the MLO image. Due to the deformable morphological characteristics of the breast, the degree of compression in different directions is different, which makes it difficult for the radiologist to compare the CC image. Linked with the MLO image; 2) Since the mammography image itself is a two-dimensional image, the image content is highly overlapping, and there is often a high degree of fusion between the calcification point or mass and the surrounding tissue image, and the image features of the calcification point or mass are not obvious or not exists, resulting in poor results of the above algorithm

Images