Method and device for classifying and locating breast lesions based on nuclear magnetic images

Abstract

The present disclosure relates to a method and device for classifying and locating breast lesions based on nuclear magnetic images. The method includes: performing two-dimensional segmentation on the first image sequence to be processed obtained by fusing multiple sets of image sequences of the target organ, Then perform three-dimensional segmentation processing, and merge the connected regions in the obtained three-dimensional segmentation results to obtain multiple first lesion sub-regions after merging; finally, according to multiple sets of image sequences of the first lesion sub-region and the target organ, Perform remerging and classification processing on the first lesion sub-region, and determine the category of the merged second lesion sub-region, and the category includes a false positive category and multiple symptom categories. According to the embodiments of the present disclosure, the lesion can be quickly and accurately identified from the breast nuclear magnetic image sequence, and the reference type of the lesion can be given, which can greatly save the doctor's viewing time and help improve the doctor's work efficiency and accuracy.

Description

technical field [0001] The present disclosure relates to the technical field of computer vision, and in particular to a method and device for classifying and locating breast lesions based on nuclear magnetic images. Background technique [0002] As one of the common malignant tumors, the incidence of breast cancer is increasing year by year, and the age of onset is getting younger and younger, seriously endangering women's physical and mental health. Early detection, early diagnosis and early treatment of breast cancer are important factors to improve prognosis, therefore, early breast cancer screening and detection are becoming more and more important. [0003] Magnetic Resonance Imaging (MRI) is an image examination method based on the principle of nuclear magnetic resonance, which has the characteristics of multiple imaging parameters and high tissue resolution. It has a good diagnostic function for brain, thyroid, liver, gallbladder, spleen, kidney, pancreas, adrenal gl...

AI Technical Summary

Problems solved by technology

However, there are many image sequences in MRI images, which will cause the doctor to take a long time to read the images

In addition, during the analysis of breast cancer image sequences, due to the diversity and complexity of breast lesions, the doctor's diagnostic level and clinical experience are relatively high, and it is difficult for junior doctors to accurately and quickly detect and analyze breast lesions

Images