Multi-modal image analysis method and system for cancer diagnosis

A cancer diagnosis, multi-modal technology, applied in image analysis, medical automation diagnosis, image enhancement and other directions, can solve the problems of poor clinical effect, low accuracy of breast cancer pathological classification evaluation, ignoring tumor heterogeneity, etc. Reduce missed diagnosis and misdiagnosis and improve screening efficiency

Pending Publication Date: 2021-03-12
ZHONGNAN HOSPITAL OF WUHAN UNIV
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the current radiomics analysis of breast cancer is combined with a single high-frequency ultrasound or X-ray molybdenum palladium or magnetic resonance (MRI) imaging examination mode, and the accuracy of breast cancer pathological classification is still low. Poor outcome, often requires further invasive histopathological examination
However, biopsy or postoperative pathological detection only takes part of the tumor sample tissue, ignoring the possible heterogeneity of the tumor, especially the larger mass tissue, which has certain limitations.

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
  • Multi-modal image analysis method and system for cancer diagnosis
  • Multi-modal image analysis method and system for cancer diagnosis
  • Multi-modal image analysis method and system for cancer diagnosis

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0070] see figure 1 and figure 2 As shown, the embodiment of the present invention provides a multimodal image analysis method for cancer diagnosis, the method comprising the following steps:

[0071] Collect the image of the lesion area, and extract the suspected lesion area through the feature pyramid network;

[0072] The false positive attenuation network based on the context feature is used to classify the suspected lesion candidate area, and the true positive ROI is extracted;

[0073] The extracted true positive ROIs were multi-scale transformed and input into the integrated neural network, feature extraction and prediction model were constructed, combined with clinical and pathological information to predict the probability of pathological type, and output the weighted average of pathological type probability in the integrated neural network.

[0074] The method of the embodiment of the present invention can be applied to the diagnosis of breast cancer, thyroid canc...

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 provides a multi-modal image analysis method and system for cancer diagnosis, and relates to the technical field of image data analysis. The method comprises the steps: extracting a suspected lesion region through a feature pyramid network, carrying out the classification through a false positive attenuation network based on context features, and extracting a true positive ROI. According to the invention, the classification neural network based on the residual network is adopted, the pathological type is stably and accurately predicted, doctors are assisted in improving the screening efficiency, missed diagnosis and misdiagnosis are effectively reduced, and medical image-based cancer diagnosis and reasonable treatment of primary hospitals are facilitated.

Description

technical field [0001] The present invention relates to the technical field of image data analysis, in particular to a multimodal image analysis method and system for cancer diagnosis. Background technique [0002] Early cancer screening and treatment has become a major health issue worldwide, and imaging examinations are the main medical means for early detection and treatment of breast cancer, thyroid cancer and other cancers. Currently commonly used imaging techniques include mammography, ultrasound, MRI, etc., which can not only realize the location of lesions but also observe and evaluate the entire tumor tissue, such as location, number, size, shape, edge shape, density, internal echo, signal or degree of enhancement. Doctors use the observation of medical imaging pictures to determine the location and judgment of cancer foci and further evaluate the pathological types of tumors as much as possible. However, in the field of medical imaging, there is a large gap betwee...

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): G16H50/20G06T7/00G06T7/11G06K9/62G06N3/04
CPCG16H50/20G06T7/0012G06T7/11G06T2207/10088G06T2207/20084G06T2207/30068G06T2207/30096G06N3/045G06F18/2415G06F18/253
Inventor 吴猛
Owner ZHONGNAN HOSPITAL OF WUHAN 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

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