A thyroid disease diagnosis method based on an SPECT image

A technology for thyroid disease and diagnostic method, applied in the field of thyroid disease diagnosis, can solve problems such as low accuracy and poor performance, achieve accurate detection results, overcome the weakening of the sum of features, and achieve great flexibility

Pending Publication Date: 2019-05-07
HARBIN INST OF TECH AT WEIHAI
View PDF2 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] Aiming at the problems of low accuracy and poor performance of the existing CNN method applied to SPECT images, the present invention proposes a method for diagnosing thyroid diseases based on SPECT images. The method adopts an improved CNN network structure. Adding trainable weight parameters to the cross-layer connection enables the network to learn weight parameters during training, which overcomes the problem that the original network's cross-layer connection will cause information redundancy and reduce network performance, thereby improving the recognition method. accuracy

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
  • A thyroid disease diagnosis method based on an SPECT image
  • A thyroid disease diagnosis method based on an SPECT image
  • A thyroid disease diagnosis method based on an SPECT image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0023] Below in conjunction with accompanying drawing, the specific embodiment of the method for diagnosing thyroid disease based on SPECT image provided by the present invention is described as follows:

[0024] figure 1 An improved network structure is given, and trainable weight parameters for cross-layer connections are added in the densely connected dense block module. figure 2 The specific network information of an implementation example is given, and the implementation of the present invention includes but is not limited to the network information.

[0025] The network in the specific embodiment of the present invention uses deep learning framework PyTorch to realize, according to figure 2 To set up and connect the layers of the network, follow the figure 1 and Equation (2) to set the way the network learns. Load the DenseNet121 provided by the framework, use the transfer learning method to pre-train the network with ImageNet, and then fine-tune it with the SPECT i...

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 thyroid disease diagnosis method based on an SPECT image. A convolutional neural network with an improved DenseNet network structure is used for image classification. Parameters influencing the weight are added to cross-layer connection in a Dense block, and the weight of the feature map of each layer is dynamically adjusted in training, so that the network has higher flexibility and the classification performance is improved. Implementations show that the method can achieve better performance than other deep learning methods. The method can be widely applied to thyroid disease diagnosis and other image classification problems.

Description

technical field [0001] The invention relates to a method for diagnosing thyroid diseases based on SPECT images. Background technique [0002] The thyroid produces thyroid hormones, which play a vital role in controlling the body's metabolism. Thyroxine (T4) and triiodothyronine (T3) are two active thyroid hormones produced by the thyroid gland that do a lot for the body, including protein production, temperature regulation, and energy production and regulation. Thyroid disease is the second largest disease in the field of endocrinology [1] , severe thyroid disease can lead to death [1-3]. [0003] There are many factors commonly used in clinical diagnosis of thyroid disease, such as clinical evaluation, blood test, imaging examination, biopsy and so on. Among them, imaging method is a very important method for thyroid diagnosis, and these images mainly include ultrasound, CT, SPECT and so on. Ultrasound is a convenient, real-time, and economical imaging method, which is ...

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): G06T7/00G06N3/04A61B6/03A61B6/00
Inventor 马立勇马城宽张湧林文靖张姝婷孙明健
Owner HARBIN INST OF TECH AT WEIHAI
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