Improved U-net kidney tumor segmentation method

A kidney tumor and kidney technology, applied in the field of tumor segmentation, can solve the problems of incomplete segmentation images and unclear boundaries.

Pending Publication Date: 2020-09-04
LIAONING TECHNICAL UNIVERSITY
View PDF7 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although the U-net model solves the problem of small data training, it still has ...

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
  • Improved U-net kidney tumor segmentation method
  • Improved U-net kidney tumor segmentation method
  • Improved U-net kidney tumor segmentation method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] The specific implementation of the present invention will be described in detail below in conjunction with the accompanying drawings. As a part of this specification, the principles of the present invention will be described through examples. Other aspects, features and advantages of the present invention will become clear through the detailed description. In the referenced drawings, the same reference numerals are used for the same or similar components in different drawings.

[0032] Deep learning has been widely used in the field of medical image semantic segmentation, because medical image semantics are relatively simple, the structure is relatively fixed and the amount of data is small. U-net combines features of different scales, and at the same time skips the connection to ensure that the features recovered by upsampling will not be very rough. Therefore, U-net is one of the most commonly used neural network models for medical image segmentation.

[0033]U-net a...

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 discloses an improved U-net kidney tumor segmentation method. The method comprises the following steps: combining a kidney and a kidney tumor together through a pixel superposition method; extracting the image for a plurality of times by using convolution at the encoder part, and generating corresponding segmented kidney and kidney tumor images by using transposed convolution at thedecoder part; introducing a residual learning unit into the encoder; and carrying out optimization training by using a Luoshi loss function in network training. According to the method, a deep residual network and a The Lovasz-Softmax loss function are combined to achieve the semantic segmentation method based on the ResUnet, the characteristics of the image are better extracted by utilizing the Lovasz hinge residual network, the improved method is more excellent than an original method, and the segmentation precision of the kidney tumor is effectively improved.

Description

technical field [0001] The invention belongs to the technical field of tumor segmentation, and in particular relates to an improved U-net kidney tumor segmentation method. Background technique [0002] Diagnosing renal tumors through imaging examinations is an important means for doctors to diagnose renal cancer. However, due to the different subjective ideas of different doctors, there are differences in the results of lesion segmentation, which will lead to unsatisfactory segmentation results and affect the judgment of the disease. Therefore, the semantic segmentation results of kidney images are of great help to the diagnosis of kidney cancer. At present, there are two ways to solve the above problems at home and abroad: on the one hand, the method based on deep learning is used; on the other hand, the traditional algorithm is used. Ren Lu et al. (Ren Lu, Li Qiang, Guan Xin, et al. Improved continuous maximum flow algorithm for three-dimensional segmentation of brain tum...

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): G06T7/11G06T7/194G06N3/04G06N3/08
CPCG06T7/11G06T7/194G06N3/08G06T2207/10081G06T2207/20081G06T2207/20084G06T2207/30084G06T2207/30096G06N3/045
Inventor 孙劲光宋晟民
Owner LIAONING TECHNICAL UNIVERSITY
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