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

A classification method for vaginal pathology images based on convolutional neural network

A convolutional neural network, pathological image technology, applied in neural learning methods, biological neural network models, image analysis, etc., to achieve the effect of low accuracy

Active Publication Date: 2022-06-07
HUNAN UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the classification of vaginal pathology images based on convolutional neural networks is still in the blank stage

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
  • A classification method for vaginal pathology images based on convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0019] The present invention is described in further detail below in conjunction with the accompanying drawings and specific embodiments:

[0020] Figure 1 The flowchart of the vaginal pathology image classification method shown contains the five steps from step one to step five, and the specific content is as follows:

[0021]Step 1: Increase the number of sub-category vaginal pathology images by upsampling method, and improve the category balance of the labeled vaginal pathology image dataset, wherein the upsampling method includes the following steps:

[0022] 1.1. Enter the vaginal pathology image dataset D= [D]. 1 ,D 2 ,D 3 ], where D represents the set of images of vaginal pathology, D 1 ，D 2 ，D 3 A subset of the vaginal pathology image datasets with categories of bacterial vaginosis negative, bacterial vaginosis intermediate, and bacterial vaginosis positive, respectively;

[0023] 1.2. Sort the original pathological image samples in order of category and calculate the num...

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 belongs to the technical field of computer vision and machine learning, and discloses a method for classifying vaginal pathological images based on a convolutional neural network. The invention comprises the steps of: using an upsampling method to obtain a category-balanced vaginal pathology image data set; using a data enhancement method to amplify the vaginal pathology image data set; using the enlarged vaginal pathology image data set to train an image classification convolutional neural network ; Use the cross-entropy loss function, combined with the BP algorithm to update the network parameters of the image classification convolutional neural network; classify the input image through the trained optimal image classification convolutional neural network. The present invention avoids the limitations of traditional feature extraction methods, such as: highly dependent on the experience and knowledge of medical staff, it takes a lot of time and energy to complete, and there are often certain difficulties in extracting distinguishable high-quality features, and the accuracy rate is low , the present invention realizes high-precision classification of vaginal pathological images by means of a convolutional neural network.

Description

Technical fields: [0001] The present invention belongs to the field of computer vision and machine learning techniques, relates to vaginal pathology image classification methods, in particular to a vaginal pathology image classification method based on convolutional neural networks. Background: [0002] With the development of science and technology and the promotion of medical imaging applications, more and more medical images need to be interpreted by doctors. Medical image interpretation has gradually become a challenging task, and doctors may have misinterpretation errors due to lack of experience or fatigue, so that some diseases are missed, resulting in false negatives, and non-lesions may be interpreted as lesions, or benign lesions may be misunderstood as malignant, resulting in false positives. In this situation, medical image recognition has become a research hotspot. [0003] Medical imaging recognition is a multidisciplinary intersection of integrated medical imaging,...

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 Patents(China)
IPC IPC(8): G06T7/00G06V10/764G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06T7/0012G06N3/084G06T2207/10004G06T2207/30096G06T2207/20081G06T2207/20084G06N3/045G06F18/2411
Inventor 彭绍亮程敏霞李非王力杨亚宁周德山李肯立毕夏安唐卓蒋洪波王树林高亦博
Owner HUNAN UNIV