Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Pathological image diagnosis of cervical cancer based on poisson 's ring conditional random field

A conditional random field, image diagnosis technology, applied in image analysis, image data processing, instruments, etc., can solve the problems of limited number of pathologists, differences in the judgment of the same pathological image, insufficient experience, etc., to improve the accuracy of diagnosis , the effect of reliable diagnostic results

Inactive Publication Date: 2019-03-01
NORTHEASTERN UNIV
View PDF4 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] 1.2 Objective Shortcomings of the Prior Art
[0010] (1) In the existing technology, the histopathological images of cervical cancer are only used as basic facts, and experienced pathologists are still required to make judgments on the images. There are also differences in judgment, which may produce large errors
[0011] (2) Existing technology still requires experienced pathologists to make judgments, but the number of pathologists is limited, and medical resources are lacking in underdeveloped areas, and pathologists are also scarce; medical students or pathologists with insufficient experience , cannot make reliable judgments on histopathological images
[0012] (3) The existing technology can only use a conditional random field-based classifier to judge whether a local area is abnormal, and cannot give the classification results of cervical cancer, such as precancerous lesions (CINI, CINII, CINIII grade 3) and malignant tumors (Three grades of high, middle and low differentiation)

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 image diagnosis of cervical cancer based on poisson 's ring conditional random field
  • Pathological image diagnosis of cervical cancer based on poisson 's ring conditional random field
  • Pathological image diagnosis of cervical cancer based on poisson 's ring conditional random field

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0062] Such as figure 2 Shown: This embodiment discloses a method for diagnosing cervical cancer histopathological images based on Poisson ring conditional random field, the method comprising:

[0063] 101. Acquiring digitized cervical cancer histopathological images.

[0064] 102. Preprocessing the acquired digitized histopathological images of cervical cancer.

[0065] 103. Using a segmentation algorithm to cluster, segment and divide the preprocessed digital cervical cancer histopathological images to obtain multiple small image blocks.

[0066] 104. Extract features from each small image block obtained in step 103, and then perform feature selection on the extracted features.

[0067] 105. Using the conditional random field to classify the selected features in the digital cervical cancer histopathological image, and obtain the grading result of the cervical cancer histopathological image.

[0068] It should be noted that the format of the digitized cervical cancer hist...

Embodiment 2

[0109] Such as image 3 Shown: this embodiment discloses a method for diagnosing cervical cancer histopathological images based on Poisson ring conditional random field, comprising the following steps:

[0110] Step 1: Collect digitized cervical cancer histopathological images for system training. The image formats include *.bmp, *.BMP, *.dip, *DIP, *.jpg, *.JPG, *.jpeg, *JPEG, * .jpe, *.JPE, *.jfif, *JFIF, *.gif, *.GIF, *.tif, *.TIF, *.tiff, *.TIFF, *.png, *.PNG, etc.: For example, this The embodiment adopts a database of 307 cervical cancer histopathological images with high, medium and low differentiation for system training (20 high, high and low differentiation images in the training set and 247 images in the test set), and the size of each image is 2560×1920 pixels.

[0111] Step 2: Preprocess the collected image: first use the median filter to denoise the image, and then use histogram equalization to enhance the image contrast. (Grayscale image available here)

[011...

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 relates to a cervical cancer tissue pathological image diagnosis method based on a Poisson ring conditional random field. The method comprises the following steps: 101, acquiring digitalcervical cancer histopathological image; 102, preprocessing the obtained digital histopathological images of cervical cancer; 103, clustering, segmenting and blocking the preprocessed digital cervical cancer histopathological images by using a segmentation algorithm, and obtaining a plurality of image blocks; 104, extracting features from each image block obtained in step 103, and then performingfeature selection on the extracted features; 105. The histopathological images of cervical cancer are graded by conditional random field model, and the graded results of histopathological images areobtained. The invention provides a cervical cancer tissue pathological image diagnosis method based on a Poisson ring conditional random field, which can obtain the grading result of cervical cancer according to the cervical cancer tissue pathological image.

Description

technical field [0001] The invention belongs to the technical field of cervical cancer histopathological image diagnosis, and in particular relates to a method for diagnosing cervical cancer histopathological images based on Poisson ring condition random field. Background technique [0002] 1.1 Brief description of the scheme of the prior art [0003] Existing techniques use conditional random fields to classify cervical colposcopy images, with histopathology images as a ground truth. Such as figure 1 As shown, the method consists of five steps from top to bottom: [0004] (1) Preprocessing includes three parts: image calibration, image registration and anatomical feature extraction; [0005] (2) Image segmentation uses a k-means-based clustering method to identify subregions in each tissue type that are homogeneous in color and intensity; [0006] (3) Diagnostically relevant features including, but not limited to, acetic acid blanching, mosaicism, punctate, and atypical...

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): G06T7/00G06T7/10
Inventor 李晨陈昊胡志杰孙洪赞张乐许宁钱唯马贺薛丹尚麟静
Owner NORTHEASTERN UNIV
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
Eureka Blog
Learn More
PatSnap group products