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

Cervical cell segmentation method based on deep learning

A cervical cell and deep learning technology, which is applied in the field of image segmentation, can solve the problems of cell edge occlusion, unsatisfactory segmentation effect, and inconspicuous background contrast, and achieve the effects of high stability, improved accuracy, and convenient and fast operation

Pending Publication Date: 2021-08-17
HARBIN UNIV OF SCI & TECH
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to solve the existing cell boundary segmentation is still the difficulty of cell segmentation, on the one hand, due to blurred cell edges, the contrast with the background is not obvious; on the other hand, due to cell overlap, the cell edge is blocked, which makes the segmentation effect not satisfactory problem; the purpose of the present invention is to provide a method for segmenting cervical cells based on deep learning

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
  • Cervical cell segmentation method based on deep learning
  • Cervical cell segmentation method based on deep learning
  • Cervical cell segmentation method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 2

[0032] Specific implementation mode two: This specific implementation mode adopts a cell segmentation method based on a generative confrontation network:

[0033] The overall scheme design of the method and figure 1 Roughly the same, the only difference is that there is an extra post-processing step after the prediction segmentation. First, preprocess the cell image, then use the generative confrontation network to segment the cervical cells, then perform post-processing, use the pit area detection algorithm to locate and segment the pit pairs of overlapping cells, and finally segment the overlapping cells. Evaluate the segmentation results.

[0034] (1) Generative confrontation network:

[0035] The GAN network is used to achieve the purpose of segmenting cervical cells in a generative manner. The GAN network uses a self-built model to distinguish the cells to be segmented and the background (image information other than the cells to be segmented), and then obtains the seg...

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 cervical cell segmentation method based on deep learning, and relates to the technical field of image segmentation. The improved cell nucleus segmentation method based on the U-Net network comprises the following steps: firstly, preprocessing cell pictures and labeled pictures, and dividing the preprocessed cell pictures and the labeled pictures corresponding to the preprocessed cell pictures into a training set, a verification set and a test set; training the network by using the training set, evaluating the trained network model by using the verification set, and storing the best model; and inputting the pictures of the test set into the model with the best evaluation result for segmentation to obtain a predicted picture, and finally evaluating the segmentation effect. According to the method, automatic segmentation of the cell nucleus is realized through the improved U-Net network model, and the accuracy can be improved; overlapped cells can be accurately and effectively segmented, rapid operation is facilitated, and the stability is high.

Description

technical field [0001] The invention belongs to the technical field of image segmentation, and in particular relates to a cervical cell segmentation method based on deep learning. Background technique [0002] Cervical cancer is the second most common cancer among Chinese women, with a high incidence rate and an increasingly younger trend. According to statistics from the World Health Organization (WHO), about 311,000 people died of cervical cancer in 2018, and there were about 570,000 new cases of cervical cancer in underdeveloped areas (accounting for 84% of the global total). There are about 150,000 new cases of cervical cancer in my country every year, accounting for about 1 / 5 of the total number of patients in the world. Nearly 80,000 women die as a result, affecting the happy life of tens of millions of families in China. However, it is worth noting that cervical cancer is a tumor with a clear cause, preventable and screenable, and it is also the only gynecological ca...

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): G06T7/12G06T7/13G06T7/181
CPCG06T7/12G06T7/13G06T7/181G06T2207/20081G06T2207/20084
Inventor 黄金杰崔鸿雁
Owner HARBIN UNIV OF SCI & TECH