Systems and methods for image processing

Patent metadata

Field	Value
Publication number	US-12380556-B2
Application number	US-202217661268-A
Country	US
Kind code	B2
Filing date	Apr 28, 2022
Priority date	Apr 28, 2021
Publication date	Aug 5, 2025
Grant date	Aug 5, 2025

Abstract

A method may include obtaining an image of an object; obtaining a target artifact identification model; determining target artifact feature information by inputting the image into the target artifact identification model, the target artifact feature information indicating a feature of one or more artifacts in the image; obtaining a target artifact extent determination model; determining target artifact extent indication information by inputting the image and the target artifact feature information into the target artifact extent determination model, the target artifact extent indication information indicating an influence extent of the one or more artifacts on an image quality of the image; in response to determining that the influence extent is greater than or equal to the extent threshold, outputting a notice of the one or more artifacts of the image.

First claim

What is claimed is:

1. A system for image processing, comprising:
at least one storage device including a set of instructions; and
at least one processor configured to communicate with the at least one storage device, wherein when executing the set of instructions, the at least one processor is configured to direct the system to perform operations including:
obtaining an image of an object;
obtaining a target artifact identification model;
determining target artifact feature information by inputting the image into the target artifact identification model, the target artifact feature information indicating a feature of one or more artifacts in the image;
obtaining a target artifact extent determination model;
determining target artifact extent indication information by inputting the image and the target artifact feature information into the target artifact extent determination model, the target artifact extent indication information indicating an influence extent of the one or more artifacts on an image quality of the image;
determining, based on the target artifact extent indication information, whether the influence extent is greater than or equal to an extent threshold; and
in response to determining that the influence extent is greater than or equal to the extent threshold, outputting a notice of the one or more artifacts of the image,
wherein the target artifact extent determination model is obtained by:
obtaining a second training sample set including a plurality of second training samples, each of the plurality of second training samples including a second training sample image and training artifact extent indication information corresponding to the second training sample image; and
obtaining the target artifact extent determination model by training, based on the second training sample set, a second preliminary model.

2. The system of claim 1 , wherein the at least one processor is configured to direct the system to perform the operations further including:
identifying, in the image, an anatomical portion of the object and an imaging plane of the object.

3. The system of claim 2 , wherein identifying the anatomical portion of the object and the imaging plane of the object before determining the target artifact feature information includes:
obtaining magnetic field intensity information of an imaging device, the imaging device being configured to scan the object to generate the image;
obtaining an imaging information identification model corresponding to the magnetic field intensity information; and
identifying the anatomical portion and the imaging plane of the object by inputting the image into the imaging information identification model.

4. The system of claim 3 , wherein the target artifact identification model is obtained based on the magnetic field intensity information of the imaging device corresponding to the image, the anatomical portion, and the imaging plane.

5. The system of claim 2 , wherein identifying the anatomical portion of the object and the imaging plane of the object after determining the target artifact feature information includes:
obtaining magnetic field intensity information of an imaging device, the imaging device being configured to scan the object to generate the image;
obtaining an imaging information identification model corresponding to the magnetic field intensity information and the target artifact feature information; and
identifying the anatomical portion and the imaging plane of the object by inputting the image into the imaging information identification model.

6. The system of claim 5 , wherein the target artifact identification model is obtained based on the magnetic field intensity information of the imaging device corresponding to the image.

7. The system of claim 2 , wherein the target artifact extent determination model is obtained based on at least one of magnetic field intensity information of an imaging device corresponding to the image, the target artifact feature information, the anatomical portion, or the imaging plane.

8. The system of claim 1 , wherein the target artifact identification model is obtained by:
obtaining a first training sample set including a plurality of first training samples, each of the plurality of first training samples including a first training sample image and training artifact feature information corresponding to the first training sample image; and
obtaining the target artifact identification model by training, based on the first training sample set, a first preliminary model.

9. The system of claim 8 , wherein obtaining the target artifact identification model by training, based on the first training sample set, the first preliminary model includes:
performing, based on a standard score, normalization on a brightness of each first training sample image in the first training sample set; and
obtaining the target artifact identification model by training, based on the normalized first training sample set, the first preliminary model.

10. The system of claim 1 , wherein the target artifact identification model is obtained based on a third training sample set, the third training sample set including third training sample images each of which is with a first label and fourth training sample images without the first label, the first label including training artifact feature information.

11. The system of claim 10 , wherein the target artifact identification model is obtained by:
obtaining a candidate artifact identification model by training, based on the third training sample images, a first preliminary model;
obtaining an intermediate artifact feature information by inputting the fourth training sample images into the candidate artifact identification model;
designating the intermediate artifact feature information as a first pseudo label of the fourth training sample images;
obtaining the target artifact identification model by training, based on the third training sample images with the first label and the fourth training sample images with the first pseudo label, the candidate artifact identification model.

12. The system of claim 11 , wherein obtaining the target artifact identification model by training, based on the third training sample images with the first label and the fourth training sample images with the first pseudo label, the candidate artifact identification model includes:
obtaining a first output artifact feature information by inputting the third training sample images into the candidate artifact identification model;
determining a first loss value based on the first output artifact feature information and the first label;
obtaining a second output artifact feature information by inputting the fourth training sample images into the candidate artifact identification model;
determining a second loss value based on the second output artifact feature information and the first pseudo label;
determining a first target loss value based on the first loss value and the second loss value; and
obtaining the target artifact identification model by updating, based on the first target loss value, the candidate artifact identification model.

13. The system of claim 12 , wherein determining the first target loss value based on the first loss value and the second loss value includes:
obtaining a first weight corresponding to the first loss value and a second weight corresponding to the second loss value; and
determining the first target loss value by determining, based on the first weight and the second weight, a weighted sum of the first loss value and the second loss value.

14. The system of claim 1 , wh

Assignees

• Shanghai United Imaging Healthcare Co Ltd

Inventors

• Xin Yang
• Shi Yuhang
• Li Guobin
• Hu Lixiang
• Chen Yanxia

Classifications

• G06T2207/30168

  Image quality inspection · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06V2201/03

  Recognition of patterns in medical or anatomical images · CPC title

• G06V10/98

  Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns · CPC title

• G06V10/82

  using neural networks · CPC title