Tumor postoperative lifetime prediction method and system based on medical image and pathological image
A technology for medical imaging and pathological images, applied in the field of image processing, can solve problems such as little-known twin network structure, achieve the effect of enhancing robustness and generalization ability, and enhancing response value
Pending Publication Date: 2021-11-23
北京知见生命科技有限公司 +1
View PDF0 Cites 3 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] Although the Siamese network is widely used in the image field, the application of the Siamese network structure in the medical field is less known
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0036] The innovation of the present invention lies in that medical images (such as CT images) and pathological images (such as HE slice images) are effectively fused together through the twin network structure, and then the fused features are used to predict postoperative tumor survival. Therefore, using the Siamese network structure to combine medical images and pathological images has broad application prospects in the medical field.
[0037] The present invention proposes a method for predicting postoperative survival of tumors combining medical images and pathological images, including:
[0038] S1. Obtain the original images of medical images and pathological images. These two types of images need to be in one-to-one correspondence and belong to the same patient. Then scale the image to a size of 512×512 pixels.
[0039] S2. Respectively transmit the scaled medical images and pathological images to the basic network to extract basic feature maps. At the same time, the ...
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
Login to View More Abstract
The invention provides a tumor postoperative lifetime prediction method and system based on medical images and pathological images. And the medical image and the pathological image are fused together through a twin network structure to predict the postoperative lifetime of the tumor. Network structures with different depths are used to extract basic features of a medical image and a pathological image, and the basic features of the medical image and the pathological image are fused together in an information interaction mode. Useless background information is effectively filtered through a channel attention network structure and a feature attention network structure, the response value of target information is enhanced, and finally a prediction result is output through a full connection layer. The whole network structure adopts an end-to-end model, so that the network can effectively learn feature information of medical images and pathological images, and robustness and generalization ability of an algorithm are enhanced.
Description
technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a twin network structure combined with medical images and pathological images for prediction of postoperative survival of tumors. Background technique [0002] In recent years, deep learning has gradually become popular in the medical field. Due to the powerful nonlinear modeling capabilities of deep learning networks and the characteristics of large amounts of information, rich features, and many types of modalities in medical images, deep learning networks are more and more widely used in medical images. [0003] The twin network structure was first used in the paper "Signature Verification using a'Siamese' Time Delay Neural Network" published on NIPS in 1993 for signature verification on American checks, that is, to verify whether the signature on the check is consistent with the signature reserved by the bank. In 2010, Hinton published the article "Recti...
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
Login to View More IPC IPC(8): G06T7/00G06N3/04G06N3/08G16H50/30G06K9/62G06T5/50
CPCG06T7/0012G06N3/084G16H50/30G06T5/50G06T2207/10081G06T2207/10024G06T2207/20221G06N3/045G06F18/253
Inventor 詹紫微张振华陈伟
Owner 北京知见生命科技有限公司