Granted Patent B2

US 12,657,800 · App. 18/488,002 · Granted Jun 16, 2026

Inventors: Meili Yang (Shanghai, CN); Yanfeng Du (Shanghai, CN); Jianwei Fu (Shanghai, CN)

Assignee: SHANGHAI UNITED IMAGING HEALTHCARE CO., LTD.

G06T11/006G06T2211/408G06T2211/441G06T2211/444

Abstract

Claims (68)

1 . A method for imaging, comprising:

obtaining a first image and topology data of an object, the first image being obtained by scanning the object under a first energy level;

determining a base material density image corresponding to the object; and

determining, based on the first image, the base material density image, and the topology data, a second image corresponding to the first image, the second image corresponding to a second energy level, and the second energy level being different from the first energy level.

2 . The method of claim 1 , wherein the second energy level exceeds the first energy level.

3 . The method of claim 1 , wherein the first energy level exceeds the second energy level.

4 . The method of claim 1 , wherein the topology data includes a difference between first topology data and second topology data of the object respectively corresponding to the first energy level and the second energy level, and the determining, based on the first image, the base material density image, and the topology data, a second image corresponding to the first image includes:

determining, based on the base material density image and the difference between first topology data and second topology data of the object, an image decomposition matrix difference;

determining, based on the base material density image and the image decomposition matrix difference, an image difference; and

determining, based on the first image and the image difference, the second image.

5 . The method of claim 4 , wherein the determining, based on the base material density image and the difference in the topology data, an image decomposition matrix difference includes:

determining, based on the base material density image, material density data; and

determining, based on the material density data and the difference between first topology data and second topology data of the object, the image decomposition matrix difference.

6 . The method of claim 4 , wherein the first topology data is obtained from scan data corresponding to the first image, or obtained by an imaging device via scanning the object under the first energy level.

7 . The method of claim 1 , wherein the determining a base material density image corresponding to the object includes:

determining, by processing the first image based on a matrix inverse decomposition or an iterative material decomposition, the base material density image.

8 . The method of claim 1 , wherein the determining a base material density image includes:

determining, by processing the first image based on a trained image processing model, the base material density image.

9 . The method of claim 8 , wherein the trained image processing model is obtained by operations, including:

obtaining a plurality of training samples; and

determining, by training a preliminary image processing model based on the plurality of training samples, the trained image processing model, wherein each training sample in the plurality of training samples includes a sample first image of a sample subject, a sample second image of the sample subject, sample topology data of the sample subject, and a label base material density image corresponding to the sample first image.

10 . The method of claim 9 , wherein the determining, by training a preliminary image processing model based on the plurality of training samples, the trained image processing model includes:

for each training sample, determining, based on the preliminary image processing model, a predicted base material density image corresponding to the sample subject; and

adjusting, based on the predicted base material density image, a parameter of the primary image processing model to optimize a value of a target loss function, and obtaining the trained image processing model, wherein the target loss function is determined based on at least one of the label base material density image, the sample first image, the sample second image, or the sample topology data.

11 . The method of claim 10 , wherein the target loss function includes at least one of a first function, a second function, a third function, or a fourth function, wherein

the first function is determined based on the label base material density image;

the second function is determined based on the sample first image;

the third function is determined based on the sample second image; and

the fourth function is determined based on the sample second image and the sample topology data.

12 . The method of claim 11 , wherein the sample topology data includes sample first topology data and sample second topology data, and the fourth function is determined according to operations including:

determining, based on the predicted base material density image, the sample first image, and the sample topology data, a sample auxiliary image; and

determining, based on the sample auxiliary image, the fourth function.

13 . The method of claim 12 , wherein the determining, based on the predicted base material density image, the sample first image, and the sample topology data, a sample auxiliary image includes:

determining, based on the predicted base material density image, sample material density data;

determining, based on the sample first topology data and the sample second topology data, a difference in the sample topology data;

determining, based on the sample material density data and the difference in the sample topology data, a sample image decomposition matrix difference;

determining, based on the sample image decomposition matrix difference and the predicted base material density image, a sample image difference; and

determining, based on the sample first image and the sample image difference, the sample auxiliary image.

14 . A system for imaging, comprising:

at least one storage device including a set of instructions; and

at least one processor in communication with the at least one storage device, wherein when executing the set of instructions, the at least one processor is configured to cause the system to:

obtain a first image and topology data of an object, the first image being obtained by scanning the object under a first energy level;

determine a base material density image corresponding to the object; and

determine, based on the first image, the base material density image, and the topology data, a second image corresponding to the first image, the second image corresponding to a second energy level, and the second energy level being different from the first energy level.

15 . The system of claim 14 , wherein the base material density image is determined by processing the first image based on a trained image processing model, and the trained image processing model is obtained by operations, including:

obtaining a plurality of training samples; and

determining, by training a preliminary image processing model based on the plurality of training samples, the trained image processing model, wherein each training sample in the plurality of training samples includes a sample first image of a sample subject, a sample second image of the sample subject, sample topology data of the sample subject, and a label base material density image corresponding to the sample first image.

16 . The system of claim 15 , wherein to determine, by training a preliminary image processing model based on the plurality of training samples, the trained image processing model, the at least one processor is further configured to cause the system to:

for each training sample, determine, based on the preliminary image processing model, a predicted base material density image corresponding to the sample subject; and

adjust, based on the predicted base material density image, a parameter of the primary image processing model to optimize a value of a target loss function, and obtain the trained image processing model, wherein the target loss function is determined based on at least one of the label base material density image, the sample first image, the sample second image, or the sample topology data.

17 . The system of claim 16 , wherein the target loss function includes at least one of a first function, a second function, a third function, or a fourth function, wherein

the first function is determined based on the label base material density image;

the second function is determined based on the sample first image;

the third function is determined based on the sample second image; and

the fourth function is determined based on the sample second image and the sample topology data.

18 . The system of claim 17 , wherein the sample topology data includes sample first topology data and sample second topology data, and the fourth function is determined according to operations including:

determining, based on the predicted base material density image, the sample first image, and the sample topology data, a sample auxiliary image; and

determining, based on the sample auxiliary image, the fourth function.

19 . The system of claim 18 , wherein to determine, based on the predicted base material density image, the sample first image, and the sample topology data, a sample auxiliary image, the at least one processor is further configured to cause the system to:

determine, based on the predicted base material density image, sample material density data;

determine, based on the sample first topology data and the sample second topology data, a difference in the sample topology data;

determine, based on the sample material density data and the difference in the sample topology data, a sample image decomposition matrix difference;

determine, based on the sample image decomposition matrix difference and the predicted base material density image, a sample image difference; and

determine, based on the sample first image and the sample image difference, the sample auxiliary image.

20 . A non-transitory computer readable medium, comprising at least one set of instructions for imaging, wherein when executed by one or more processors of a computing device, the at least one set of instructions causes the computing device to perform a method, the method comprising:

obtaining a first image and topology data of an object, the first image being obtained by scanning the object under a first energy level;

determining a base material density image corresponding to the object; and

determining, based on the first image, the base material density image, and the topology data, a second image corresponding to the first image, the second image corresponding to a second energy level, and the second energy level being different from the first energy level.

Assignments (1)

Priority Claims (2)

Continuity (2)

References Cited (28)