Pathological image classification method, device, equipment and storage medium for membranous nephropathy

Abstract

The present application discloses a pathological image classification method, device, equipment and storage medium for membranous nephropathy. The method includes: acquiring a renal biopsy glomerular pathological image; inputting the renal biopsy glomerular pathological image into a pre-trained The segmentation module of the MTLSC-Net neural network model obtains the segmentation mask image of the glomerular basement membrane and its surrounding immune complexes; the segmentation mask image is input into the classification module of the MTLSC-Net neural network model to obtain the membrane Classification results of pathological images of nephropathy. According to the pathological image classification method provided by the embodiment of the present application, based on the multi-task deep learning neural network model, the features related to the classification of the pathological images of membranous nephropathy are obtained by segmentation, and the pathological images of membranous nephropathy are quickly and accurately classified, so as to assist the pathologist to classify the pathological images of membranous nephropathy. The disease stage of membranous nephropathy is classified, which effectively improves the work efficiency of pathologists.

Description

technical field [0001] The invention relates to the technical field of image classification, in particular to a pathological image classification method, device, equipment and storage medium for membranous nephropathy. Background technique [0002] Membranous nephropathy is the second most common disease in nephrotic syndrome in China, and its incidence is on the rise, and nearly half of the patients may develop end-stage renal failure. Membranous nephropathy requires pathological examinations such as light microscopy, electron microscopy, and immunofluorescence through renal biopsy for diagnosis and staging. Homogeneous thickening and spike formation. The pathological staging of membranous nephropathy is usually divided into five stages, but the staging of I, II, III, and IV is more important. When staging membranous nephropathy, pathologists need to carefully observe various stained renal biopsy histological sections under electron microscope or light microscope, and con...

AI Technical Summary

Problems solved by technology

[0003] At present, various medical imaging data are increasing by orders of magnitude. Medical pathological images are complex and the tissues and cells to be detected are difficult to distinguish, which brings great challenges to traditional medical imaging pattern recognition technology.

Images