Method of detecting and identifying lesions of gastrointestinal endoscope, based on sliding window

A sliding window and recognition method technology, applied in the field of medical image intelligent processing, can solve problems such as diagnostic differences and omissions, and achieve the effect of eliminating interference

Active Publication Date: 2018-11-06
FUDAN UNIV
View PDF4 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the traditional diagnosis method, the doctor's diagnosis is a process of subjective judgment, so it will be limited and affected by the experience and knowledge level of the diagnosing doctor...

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
  • Method of detecting and identifying lesions of gastrointestinal endoscope, based on sliding window
  • Method of detecting and identifying lesions of gastrointestinal endoscope, based on sliding window
  • Method of detecting and identifying lesions of gastrointestinal endoscope, based on sliding window

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The technical solutions of the present invention will be described in detail below, but the protection scope of the present invention is not limited to the embodiments.

[0036] Use the images marked by doctors to make samples, and train the classifier until it converges. The specific implementation method is:

[0037] (1) Input the image to be tested of H×W, calculate base=min(H,W), the length and width of the candidate frame have three kinds of base×0.375, base×0.55, base×0.725, so constitute 3×3=9 kinds Combination, that is, 9 kinds of candidate boxes; slide the candidate box in the image, the vertical sliding step is H×S, the horizontal sliding step is W×S, and the parameter S of the sliding step is set to 0.04;

[0038] (2) All candidate frames are input to the classifier, and the classifier outputs the probability of candidate frame lesions;

[0039] (3) Use the threshold T to filter the candidate frame with a low lesion probability; when the lesion probability o...

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 intelligent processing of medical images, and specifically relates to a method of detecting and identifying lesions of a gastrointestinal endoscope, based on a sliding window. Early screening through endoscopy is an effective means of reducing the morbidity and mortality of cancers of the digestive tract. In traditional diagnostic methods, the diagnosis of a doctor is entirely a subjective judgment process, so that diagnosis is limited and influenced by the experience and knowledge level of the diagnostic doctor. Therefore, the method of detecting and identifying lesions of a gastrointestinal endoscope, based on a sliding window includes the steps: applying deep learning to lesion detection of a gastrointestinal endoscope, making a sample based on the frame of a lesion area marked by a doctor, and training a classifier; and proposing a candidate area in a gastrointestinal endoscope image to be detected, inputting the candidate area into the classifier, and performing post-processing on the classification result to achieve the purpose of lesion detection. The experiment result shows that the method of detecting and identifying lesionsof a gastrointestinal endoscope, based on a sliding window can accurately detect the lesion position of the gastrointestinal endoscope image so as to provide a reference for the doctor, thus showing that the artificial intelligence-assisted diagnosis and treatment of early digestive tract cancer has irreplaceable superiority.

Description

technical field [0001] The invention belongs to the technical field of medical image intelligent processing, and in particular relates to a method for detecting and identifying gastrointestinal endoscopic lesions based on a sliding window. Background technique [0002] Standardized gastrointestinal cancer screening, treatment, and follow-up are of great significance. Early cancer screening is an effective means to reduce cancer morbidity and mortality. The screening, treatment and follow-up of early gastrointestinal cancer mainly include endoscopy and postoperative CT examination, among which gastrointestinal endoscopy is the most important. In the traditional diagnosis method, the doctor's diagnosis is a process of subjective judgment, so it will be limited and affected by the experience and knowledge level of the diagnosing doctor; secondly, it is easy for the doctor to miss some subtle changes in the diagnosis; thirdly, the difference between different doctors and the sam...

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): G06T7/00
CPCG06T7/0012G06T2207/30004G06T2207/10068G06T2207/20081
Inventor 钟芸诗颜波蔡世伦牛雪静李冰林楚铭谭伟敏
Owner FUDAN 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
Try Eureka
PatSnap group products