Unlock instant, AI-driven research and patent intelligence for your innovation.

Pathological section tissue region recognition system based on image semantic segmentation

A semantic segmentation and recognition system technology, applied in the field of machine learning, can solve problems such as intricate content shape, uneven quality of pathological slices, and difficulty in accurate identification of tissue regions, achieving the effect of precise tissue region recognition

Pending Publication Date: 2020-05-19
苏州优纳医疗器械有限公司
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the uneven quality and intricate content and morphology of pathological slides, it is still very difficult to accurately identify tissue regions within slices

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
  • Pathological section tissue region recognition system based on image semantic segmentation
  • Pathological section tissue region recognition system based on image semantic segmentation
  • Pathological section tissue region recognition system based on image semantic segmentation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0042] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings.

[0043] like figure 1 , 2 As shown, the present invention comprises the following steps:

[0044] 1. Data collection and labeling

[0045] In this protocol, about 100,000 sliced ​​macroscopic images were collected, and the tissue regions in each image were marked with polygons. Some example images with corresponding annotations such as figure 1 Data collection callouts in . 80% of the collected macro images are about 80,000 as the training dataset, and 20% are about 20,000 as the test dataset.

[0046] Second, image semantic segmentation network training.

[0047] Common image semantic segmentation networks, such as DeepLab series, UNet, SegNet, FCN (FullyConvolutional Neural Networks) series, are suitable for this scheme. In order to improve the efficien...

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 a pathological section tissue region identification system based on image semantic segmentation, which realizes accurate identification of a tissue region by utilizing big dataand a deep learning technology, and the basic content of the scheme comprises the following steps: (1) data collection and labeling; (2) training of an image semantic segmentation network; (3) imagesemantic segmentation network prediction; and (4) post-processing and outputting of tissue area identification. According to the method, under the condition that the efficiency of the semantic segmentation network is improved, good convergence can still be achieved on the organization region segmentation problem, and organization region segmentation is effectively achieved.

Description

technical field [0001] The invention relates to the technical field of machine learning, in particular to a pathological slice tissue region identification system based on image semantic segmentation. Background technique [0002] The identification of tissue regions in pathological slices is a key basic technology in the application of digital pathology. Digital pathology is mainly composed of digital slice scanning device and data processing software. First, use a digital microscope or a magnifying system to scan and image the pathological slices one by one under a low-magnification objective lens. Then, on the basis of the effective enlargement of the optical magnifying device, the scanning control software uses the program-controlled scanning method to collect high-resolution digital images, and the image compression and storage software automatically stitches the images seamlessly to produce a digital slice (Whole full field of view). Slide Image, referred to as WSI)....

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/00G06T7/11G06T7/187G06T7/155
CPCG06T7/0012G06T7/11G06T7/187G06T7/155G06T2207/20081G06T2207/20084G06T2207/30096
Inventor 杨永全郑众喜袁勇雷雪梅蔡小玲王杰
Owner 苏州优纳医疗器械有限公司