Animal lesion liver pathology image recognition method based on deep learning

A pathological image and liver technology, applied in the field of image recognition, can solve the problems of inability to provide reference for liver disease staging and grading and prognosis prediction, missed judgment, wrong judgment, time-consuming and labor-intensive, etc., to improve migration and generalization ability and high accuracy. The effect of identifying and reducing the misdiagnosis rate

Inactive Publication Date: 2022-05-06
上海云纹生物科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Manually analyzing and judging fibrosis and tumors in animal liver pathological images is not only a time-consuming and labor-intensive process, but also prone to missed judgments and misjudgments. The influence of subjective and objective factors may give different judgment results
Artificial intelligence and deep learning algorithms ca...

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Animal lesion liver pathology image recognition method based on deep learning
  • Animal lesion liver pathology image recognition method based on deep learning
  • Animal lesion liver pathology image recognition method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0029] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0030] Such as figure 1 Shown, the present invention comprises the following steps

[0031] Step 1: Establish an animal lesion liver pathology dataset

[0032] 1-1: Collect pathological images of liver lesions in animals, and the pathologist manually labels each pathological image through the ASAP data labeling software. The manual labeling includes the labeling of fibrosis areas and tumor areas, and each pathological image gets a corresponding Annotate images.

[0033] 1-2: Classify all animal lesion liver pathology images according to the labeled images, and establish animal lesion liver pathology data sets, including fibrosis pathology subsets and tumor pathology subsets, wherein, if the same pathology image contains both fibrosis and tumor , then the pathological image is classified into the fibrosis pathology subset and the tumor pathology subse...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses an animal lesion liver pathology image recognition method based on deep learning, and the method comprises the steps: collecting an animal lesion liver pathology image, and building an animal lesion liver pathology data set; extracting a liver tissue area and an edge contour; improving a U-net network, training the improved U-net on a training set and a test set to obtain a tissue fibrosis segmentation model, identifying an original image in a fibrosis pathology subset, and calculating a fibrosis proportion; an NCRF network is improved, the improved NCRF is trained on a training set and a test set, a tumor segmentation model is obtained, original images in a tumor pathology subset are recognized, and the number and proportion of tumors are calculated. According to the method, different deep learning networks are adopted for different morphological characteristics of fibrosis and tumors in the pathological image of the animal lesion liver, so that the pathological analysis capability of the lesion liver is improved, and reference is provided for stage grading and prognosis prediction of the lesion liver.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a deep learning-based recognition method for pathological images of animal pathological lesions and livers. Background technique [0002] As the most important organ to maintain homeostasis, the liver undertakes most of the metabolic functions of the human body, and plays a key role in the metabolism of glucose, lipids, amino acids, heterologous organisms, and production of proteins, coagulation factors, and bile. How to effectively treat liver disease has always been a major problem plaguing human health. Normally, the liver has the ability to regenerate and repair after being subjected to certain physical and chemical damage, but excessive or continuous damage will break the balance between liver tissue damage and repair, causing irreversible liver damage, leading to liver fibrosis, Cirrhosis can even lead to hepatocellular carcinoma. Therefore, the diagnosis of liv...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06T7/00G06T7/10G06T7/13G06V10/774G06K9/62
CPCG06T7/0012G06T7/10G06T7/13G06T2207/30056G06F18/214
Inventor 于观贞
Owner 上海云纹生物科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap