Low-dose PET image restoration method and system, device, and medium

Patent metadata

Field	Value
Publication number	US-12412369-B2
Application number	US-202018265685-A
Country	US
Kind code	B2
Filing date	Dec 8, 2020
Priority date	Dec 8, 2020
Publication date	Sep 9, 2025
Grant date	Sep 9, 2025

Abstract

A low-dose PET image restoration method and system, a device, and a medium are provided. The method includes: S1, performing blocking processing on a training image comprising a low-dose PET image, an MR image, and a standard-dose PET image to obtain a first patch, and performing first preprocessing on the first patch to obtain a second patch; S2, obtaining, according to the second patch, a first joint dictionary by means of sparse coding and dictionary updating; and S3, restoring the low-dose PET image to a restored image of the standard-dose PET image according to the first joint dictionary.

First claim

What is claimed is:

1. A low-dose Positron Emission Tomography (PET) image restoration method, comprising the following steps:
S1, performing blocking processing on training images comprising a low-dose PET image, a Magnetic Resonance (MR) image, and a standard-dose PET image to obtain a first patch, and performing first preprocessing on the first patch to obtain a second patch;
S2, obtaining a first joint dictionary by sparse coding and dictionary updating according to the second patch; and
S3, restoring the low-dose PET image to a standard-dose PET restored image according to the first joint dictionary;
wherein the low-dose PET image restoration method, wherein the S2 comprises the following steps:
S21, obtaining an initialized dictionary comprising a low-dose PET dictionary, an MR dictionary, and a standard-dose PET dictionary by taking the second patch as a sample;
S22, constructing an initialized joint dictionary according to the initialized dictionary, and constructing a target matrix according to the second patch; and
S23, obtaining a sparse code according to the target matrix, iteratively updating the sparse code and the initialized joint dictionary until an iteration stop condition is met, and obtaining the first joint dictionary comprising a first low-dose PET dictionary, a first MR dictionary, and a first standard-dose PET dictionary.

2. The low-dose PET image restoration method according to claim 1 , wherein in the S23, each iteration comprises first updating the sparse code with the dictionary fixed, and then updating the dictionary with the sparse code fixed.

3. The low-dose PET image restoration method according to claim 1 , wherein in the S23, partial samples are randomly selected for the sparse coding and the dictionary updating in each iteration.

4. The low-dose PET image restoration method according to claim 1 , wherein the S3 comprises the following steps:
S31, performing blocking processing on the low-dose PET image and the MR image to obtain a third patch, and performing second preprocessing on the third patch to obtain a fourth patch;
S32, merging the first low-dose PET dictionary and the first MR dictionary obtained in the S2 into a second joint dictionary, and obtaining a second sparse code according to the fourth patch and the second joint dictionary; and
S33, obtaining a prediction patch according to the first standard-dose PET dictionary and the second sparse code obtained in the S2, and restoring the prediction patch to a two-dimensional lattice to obtain the standard-dose PET restored image.

5. The low-dose PET image restoration method according to claim 4 , wherein the third patch, at a selected position, covers an entire image in an order of a multi-frame image.

6. The low-dose PET image restoration method according to claim 4 , wherein in the S31, the second preprocessing comprises mapping a third patch of the low-dose PET image and a third patch of the MR image to imaging space of the standard-dose PET image through a preset matrix to obtain a fourth patch of the low-dose PET image and a fourth patch of the MR image.

7. The low-dose PET image restoration method according to claim 1 , wherein the first patch is a one-dimensional vector generated by randomly selecting an image block from a multi-frame image and extending the image block, and the first patch comprises a first patch of the low-dose PET image, a first patch of the MR image, and a first patch of the standard-dose PET image;
the first patch is at a same position of the multi-frame image.

8. The low-dose PET image restoration method according to claim 1 , wherein in the S1, the first preprocessing comprises mapping a first patch of the low-dose PET image and a first patch of the MR image to imaging space of the standard-dose PET image through a preset matrix to obtain a second patch of the low-dose PET image and a second patch of the MR image.

9. The low-dose PET image restoration method according to claim 1 , wherein the S21 comprises using a K-means clustering algorithm to obtain K clustering centers as the initialized dictionary by taking the second patch as the sample, and performing normalization processing on the initialized dictionary.

10. The low-dose PET image restoration method according to claim 1 , wherein in the S22, expressions of the initialized joint dictionary and the target matrix are respectively:

D
=

[

Dl

Dr

Ds

]

,

Y
=

[

Yl

Yr

Ys

]

wherein D represents the initialized joint dictionary, Y represents the target matrix, Dl represents the low-dose PET dictionary, Dr represents the MR dictionary, Ds represents the standard-dose PET dictionary, Yl represents a second patch of the low-dose PET image, Yr represents a second patch of the MR image, and Ys represents a second patch of the standard-dose PET image.

11. The low-dose PET image restoration method according to claim 1 , wherein in the S23, an expression of the sparse coding comprises:

argmin
X

⁢



Y
-

D
⁢

Λ

-
1

⁢
Λ
⁢
X


F
2

+

λ
⁢



Λ
⁢
X



1

wherein X represents sparse coding, Λ is a diagonal matrix, and a diagonal element of the diagonal matrix is A q =d q −y i , wherein da is a q th atom of a dictionary D, y i is an i th element of Y, and λ represents a sparse constraint coefficient.

12. The low-dose PET image restoration method according to claim 1 , wherein in the S23, an expression of the dictiona

Assignees

• Shenzhen Inst Adv Tech

Inventors

• Hu Zhanli
• Zheng Hairong
• Liang Dong
• Yang Yongfeng
• Liu Xin
• Xu Yingjie

Classifications

• G06T12/30

  Image post-processing, e.g. metal artefact correction · CPC title

• G06T2207/10104

  Positron emission tomography [PET] · CPC title

• G06T2207/10088

  Magnetic resonance imaging [MRI] · CPC title

• G06T7/0012

  Biomedical image inspection · CPC title

• G06T2211/464

  Dual or multimodal imaging, i.e. combining two or more imaging modalities · CPC title