Method and system for identifying renal tubular atrophy area based on deep learning

Abstract

The invention belongs to the technical field of artificial intelligence assisted medical treatment, and particularly relates to a method and a system for identifying a renal tubular atrophy area based on deep learning, and the method comprises the following steps: S1, obtaining an image of a kidney pathological section subjected to renal tubular atrophy area labeling; S2, training an instance segmentation network based on the image and a corresponding renal tubular atrophy area label, wherein the instance segmentation network is an improved mask-RCNN, a cascade network is added to the network for cascading each detection model, an IOU threshold value is set to define a sample training model, the output of the previous detection model is used as the input of one detection model, and the IOU value rises all the time; S3, inputting a to-be-detected image into the trained instance segmentation network to obtain a target frame position of each renal tubular atrophy area in the to-be-detected image; and S4, calculating the area and proportion of each renal tubular atrophy area. According to the invention, the detection of the renal tubular atrophy area can be realized, the detection precision is high, and the omission ratio is low.

Description

technical field [0001] The invention belongs to the technical field of artificial intelligence-assisted medical treatment, and in particular relates to a method and system for identifying renal tubular atrophy regions based on deep learning. Background technique [0002] Chronic kidney disease (CKD) is currently recognized by the World Health Organization (WHO) as one of the major public health challenges, and renal biopsy plays a key role in the diagnosis and treatment of CKD. At present, the pathological diagnosis of renal biopsy has become an important reference for the diagnosis, treatment and prognosis of patients with renal diseases. [0003] Pathology plays a vital role in the medical field. Among them, in the diagnosis of chronic kidney disease, the diagnostic results of the kidney tissue after coarse needle puncture, forceps, cutting, and resection are the most accurate and reliable after being made into pathological sections. "gold standard". Many kidney diseases...

AI Technical Summary

Problems solved by technology

Renal tubular atrophy is the main form of chronic kidney disease. When checking whether a patient suffers from renal tubular atrophy, a pathologist is required to observe the patient's kidney pathology section, and an authoritative pathologist is required to judge the area ratio of the atrophic area. However, medical needs and resources The contradiction between the traditional medical methods can not meet the needs of patients

The pathology departments of some tertiary hospitals produce thousands of kidney pathology sections every day. Although most of them may not have positive results, doctors still need to strictly examine each pathology section under a microscope to determine whether there is atrophy in the renal tubules, which consumes a lot of energy.

Images