Early cancer lesion range prediction auxiliary system based on deep learning

A deep learning and auxiliary system technology, applied in the field of image processing, can solve problems such as prone to errors and low accuracy of direct diagnosis, and achieve the effects of reducing workload, improving recognition accuracy, and improving processing capacity

Active Publication Date: 2019-10-22
CHONGQING UNIV CANCER HOSPITAL
View PDF6 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the accuracy of direct diagnosis in the initial stage of the current deep learnin...

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
  • Early cancer lesion range prediction auxiliary system based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0038] Such as figure 1 As shown, a deep learning-based auxiliary system for predicting the range of early cancer lesions includes an image acquisition module, a model building module, a range division module, and a display module.

[0039] The image acquisition module is used to obtain the sample image of the digestive tract endoscope with the target frame, preprocess the sample image, record the coordinate information of the end point of the target frame, sort them randomly, generate a training image set, and perform all Sample images are normalized. In this implementation, the normalization process refers to normalizing the sample image into DICOM format, NIfTI format or original binary format. The target frame is a rectangular frame, and the number of target frame end points is four.

[0040] In this embodiment, the sample image includes one or more cancer categories in early esophageal cancer, early gastric cancer, and early colon cancer, and each cancer category corres...

Embodiment 2

[0057] An auxiliary system for predicting the range of early cancer lesions based on deep learning differs from Embodiment 1 in that when the construction unit outputs a convolutional neural network model, the image acquisition module marks the adjusted preprocessing method as an effective preprocessing method. The image acquisition module is also used to acquire the images to be diagnosed from the gastrointestinal endoscope, preprocess the images to be diagnosed by an effective preprocessing method, and send the preprocessed images to be diagnosed to the range division module.

[0058] The image to be diagnosed is preprocessed through an effective preprocessing method, so that the image to be diagnosed can meet the input requirements of the convolutional neural network model, and the recognition accuracy can be improved.

Embodiment 3

[0060] An auxiliary system for predicting the range of early cancer lesions based on deep learning, the difference from Embodiment 2 is that it also includes a judgment module and an environment regulation module;

[0061] The evaluation module is used to obtain the image to be diagnosed with the target frame drawn by the doctor, compare the target frame drawn by the doctor in the same image to be diagnosed with the target frame drawn by the range division module, and judge whether they are consistent. If not, the evaluation module outputs request evaluation information ; At present, the convolutional neural network model is still mainly used for auxiliary diagnosis, and the probability of misdiagnosis is reduced through double verification, while improving the diagnostic efficiency of doctors.

[0062] The judging module is also used to receive judged information; in this embodiment, the judged information is that the doctor is correct or the convolutional neural network model...

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 the technical field of image processing, and particularly discloses an early cancer lesion range prediction auxiliary system based on deep learning. The system comprises an image acquisition module used for acquiring a sample image of a digestive tract endoscope with a target frame, preprocessing the sample image, recording coordinate information of an end point of the target frame and generating a training image set; a model construction module used for constructing a convolutional neural network model, carrying out iterative training on the convolutional neural network model based on the training image set, then carrying out testing, and obtaining a successfully trained convolutional neural network model after the testing is completed; and a range division moduleused for receiving a to-be-diagnosed image of the digestive tract endoscope, judging the to-be-diagnosed image based on the successfully trained convolutional neural network model, and outputting coordinate information of a prediction endpoint of the to-be-diagnosed image. The range division module draws a target box on the to-be-diagnosed image based on the coordinate information of the prediction endpoint. By adopting the technical scheme, the early cancer lesion range can be marked.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to an auxiliary system for predicting the range of early cancer lesions based on deep learning. Background technique [0002] Early detection and early treatment of digestive tract cancer have very important practical significance. However, the distribution of gas in the digestive tract is more, the lesions are smaller and more occult, and the morphology, surface microstructure, and surface microvessels of early cancer are very similar to inflammation and repair, which makes the pathological characteristics of digestive tract cancer complicated and difficult. Identification, early symptoms are not easy to be found. [0003] At present, the diagnosis of digestive tract cancer mainly collects images in the human body through the optical lens and image sensor of the digestive endoscope, and transmits the collected images to the display terminal for viewing by medical ...

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): G06T7/00G06K9/62
CPCG06T7/0012G06T2207/30096G06T2207/20081G06T2207/20084G06F18/241G06F18/214
Inventor 陈伟庆柴毅
Owner CHONGQING UNIV CANCER HOSPITAL
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap