Title:
PET IMAGE RECONSTRUCTION ALGORITHM COMBINING FILTERED BACK-PROJECTION ALGORITHM AND NEURAL NETWORK
Document Type and Number:
WIPO Patent Application WO/2021/232653
Kind Code:
A1
Abstract:
A PET image reconstruction algorithm combining a filtered back-projection algorithm and a neural network. The PET image reconstruction algorithm comprises the following steps: step 1, injecting a PET radioactive tracer into a biological tissue, performing scanning using a PET device, and detecting coincidence photons and counting same, so as to obtain an original sinogram matrix Y; step 2, establishing a measurement equation model according to a PET imaging principle; step 3, splitting a reconstruction problem into a reconstruction sub-problem and a de-noising sub-problem; step 4, solving the reconstruction sub-problem by using a filtered back-projection (FBP) layer, and solving the de-noising sub-problem by using an improved de-noising convolutional neural network (DnCNN); step 5, at a training stage, adjusting a parameter of an FBP-Net, and reducing an error between an output of the FBP-Net and a tag; and step 6, at an estimation stage, inputting a sinogram to be reconstructed into a trained FBP-Net, so as to directly obtain a required reconstructed image. By means of the method, the problem of it being difficult to interpret when deep learning is used for image reconstruction is solved, and a clear PET image may still be reconstructed where the count rate is low.
Inventors:
LIU HUAFENG (CN)
WANG BO (CN)
WANG BO (CN)
Application Number:
PCT/CN2020/117949
Publication Date:
November 25, 2021
Filing Date:
September 25, 2020
Export Citation:
Assignee:
UNIV ZHEJIANG (CN)
International Classes:
G06N3/04; G06T11/00; G06N3/08; G06T5/00
Foreign References:
| CN111627082A | 2020-09-04 | |||
| CN110221346A | 2019-09-10 | |||
| US20200082507A1 | 2020-03-12 | |||
| US20190266728A1 | 2019-08-29 |
Other References:
ZHANG YUNGANG, YI BENSHUN;WU CHENYUE;FENG YU: "Low- Dose CT Image Denoising Method Based on Convolutional Neural Network", ACTA OPTICA SINICA, vol. 38, no. 4, 30 April 2018 (2018-04-30), pages 123 - 129, XP055873531, DOI: 10.3788/AOS201838.0410003
GAO JINGZHI, LIU YI;ZHANG QUAN;GUI ZHIGUO: "Improved Deep Residual Convolutional Neural Network for LDCT Image Estimation", COMPUTER ENGINEERING AND APPLICATIONS, HUABEI JISUAN JISHU YANJIUSUO, CN, vol. 54, no. 16, 15 August 2018 (2018-08-15), CN , XP055873532, ISSN: 1002-8331, DOI: 10.3778/j.issn.1002-8331.1802-0055
GAO JINGZHI, LIU YI;ZHANG QUAN;GUI ZHIGUO: "Improved Deep Residual Convolutional Neural Network for LDCT Image Estimation", COMPUTER ENGINEERING AND APPLICATIONS, HUABEI JISUAN JISHU YANJIUSUO, CN, vol. 54, no. 16, 15 August 2018 (2018-08-15), CN , XP055873532, ISSN: 1002-8331, DOI: 10.3778/j.issn.1002-8331.1802-0055
Attorney, Agent or Firm:
SHENZHEN DISI-ELITE INTELLECTUAL PROPERTY AGENCY (CN)
Download PDF: