Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
DARTS NETWORK-BASED MULTI-MODAL MEDICAL IMAGE FUSION METHOD
Document Type and Number:
WIPO Patent Application WO/2022/121100
Kind Code:
A1
Abstract:
The present invention relates to the technical field of image fusion in image processing and analysis. Disclosed is a differentiable architecture search (DARTS) network-based multi-modal medical image fusion method. In the present invention, feature extraction is performed on multi-modal medical images by using a DARTS network, and the network performs learning by using a gradient of a network weight as a loss function during a search stage, and selects, between different convolution operations and connections of different nodes, a network structure that is most suitable for the current data set, so that feature details extracted by the network are more abundant; moreover, in the present invention, various indicators that can characterize image grayscale information, correlations, detail information, structural features, and image contrast are used as the network loss function, and effective fusion of medical images can be achieved in an unsupervised learning manner without a gold standard; thus, the present invention has a better fusion effect in image detail protection and image contrast enhancement compared with existing methods.

More Like This:
JP7458769Image processing device, imaging device, image processing method, program and recording medium
JP4144258Image output device and image output method
JP2002504843[Title of Invention] Breast projection imaging device
Inventors:
ZHANG XUMING (CN)
YE SHAOZHUANG (CN)
Application Number:
PCT/CN2021/074531
Publication Date:
June 16, 2022
Filing Date:
January 30, 2021
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
UNIV HUAZHONG SCIENCE TECH (CN)
International Classes:
G06T5/50
Foreign References:
CN111553480A2020-08-18
CN110852168A2020-02-28
CN111882514A2020-11-03
CN111612754A2020-09-01
US9922272B22018-03-20
Other References:
GAO YUAN;WU FAN;QIN PINLE;WANG LIFANG: "Medical Image Fusion Algorithm Based on Generative Adversarial Residual Network", JOURNAL OF COMPUTER APPLICATIONS, vol. 39, no. 12, 25 July 2019 (2019-07-25), pages 3528 - 3534, XP055941268, ISSN: 1001-9081, DOI: 10.11772/j.issn.1001-9081.2019050937
Attorney, Agent or Firm:
HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY PATENT AGENCY CENTER (CN)
Download PDF:
View/Download PDF      PDF Help



 
Previous Patent: THIN FILM PRESSURE SENSOR AND LAYOUT METHOD THEREOF

Next Patent: METHOD FOR SEPARATING DYE-CONTAINING WASTEWATER BASED ON POROUS POLYMER-MODIFIED METAL CARBON NANOTU...