Medical image processing method, medical image processing apparatus, and computer readable non-volatile storage medium storing medical image processing program

Patent metadata

Field	Value
Publication number	US-12243127-B2
Application number	US-202217699008-A
Country	US
Kind code	B2
Filing date	Mar 18, 2022
Priority date	Apr 7, 2021
Publication date	Mar 4, 2025
Grant date	Mar 4, 2025

Abstract

A medical image processing method includes obtaining a first set of projection data by performing, with a first CT apparatus including a first detector with a first pixel size, a first CT scan of an object in a first imaging region of the first detector; obtaining a first CT image with a first resolution by reconstructing the first set of projection data; obtaining a processed CT image with a resolution higher than the first resolution by applying a machine-learning model for resolution enhancement to the first CT image; and displaying the processed CT image or outputting the processed CT image for analysis. The machine-learning model is obtained by training using a second CT image based on a second set of projection data acquired by a second CT scan of the object in a second imaging region with a second CT apparatus including a second detector with a second pixel size.

First claim

The invention claimed is:

1. A medical image processing method, comprising:
obtaining a first set of projection data by performing, with a first computed tomography (CT) apparatus comprising a first detector with a first pixel size, a first CT scan of an object in a first imaging region of the first detector;
obtaining a first CT image with a first resolution by reconstructing the first set of projection data by modifying the first pixel size to a second pixel size, wherein the first CT image is used as an input image for a machine-learning model;
obtaining a processed CT image with a resolution higher than the first resolution by applying the machine-learning model for resolution enhancement to the first CT image; and
displaying the processed CT image or outputting the processed CT image for analysis, wherein
the machine-learning model is obtained by training using a second CT image generated based on a second set of projection data which is acquired by performing, with a second CT apparatus comprising a second detector with the second pixel size smaller than the first pixel size, a second CT scan of the object in a second imaging region of the second detector, the second imaging region being smaller than the first imaging region, wherein the second CT image is used as a target image for the input image.

2. The medical image processing method of claim 1 , further comprising:
in applying the machine-learning model, generating the first CT image by reconstructing the first set of projection data according to a first matrix size; and
in not applying the machine-learning model, generating another CT image by reconstructing the first set of projection data according to a second matrix size smaller than the first matrix size.

3. The medical image processing method of claim 1 , further comprising:
in applying the machine-learning model, generating the first CT image by reconstructing the first set of projection data by a first reconstruction function; and
in applying a second machine-learning model for noise reduction different from the machine-learning model in place of the machine-learning model, generating another CT image by reconstructing the first set of projection data by a second reconstruction function having a larger noise reduction effect than the first reconstruction function, and applying the second machine-learning model to the another CT image.

4. The medical image processing method of claim 3 , further comprising:
in applying the machine-learning model,
generating a plurality of 3D partial images based on the first CT image;
inputting the plurality of 3D partial images to a designated one of the machine-learning model and the second machine-learning model to obtain a plurality of processed 3D partial images by applying the designated machine-learning model, and
obtaining the processed image by combining the plurality of processed 3D partial images together.

5. The medical image processing method of claim 4 , wherein
in the generating the plurality of 3D partial images, at least two of the plurality of 3D partial images are generated in a partially overlapping manner.

6. The medical image processing method of claim 4 , wherein
in combining the plurality of processed 3D partial images, the plurality of processed 3D partial images is combined by applying filtering to a joint part between two adjacent processed 3D partial images of the plurality of processed 3D partial images.

7. The medical image processing method of claim 1 , wherein
in the obtaining the processed CT image, the processed CT image is obtained by combining, at a predetermined ratio, the first CT image and an image obtained by applying the machine-learning model to the first CT image.

8. The medical image processing method of claim 7 , wherein
the predetermined ratio is set according to a user input or a set of imaging conditions.

9. The medical image processing method of claim 1 , wherein
the machine-learning model is for applying super resolution processing to the first CT image.

10. The medical image processing method of claim 1 , wherein
the machine-learning model is for applying super resolution processing and noise reduction processing to the first CT image.

11. The medical image processing method of claim 1 , wherein
in obtaining the machine-learning model, the machine-learning model is trained with training images being the second CT image and a third CT image generated based on either the second CT image or the second set of projection data, the third CT image having a lower resolution and greater noise than the second CT image.

12. The medical image processing method of claim 11 , wherein
in obtaining the machine-learning model, the machine-learning model is trained with training images being the second CT image and a fourth CT image generated based on a third set of projection data, the third set of projection data being obtained by applying noise addition and resolution-lowering processing to the second set of projection data.

13. A medical image processing apparatus, comprising
processing circuitry configured to:
obtain a first set of projection data by performing, with a first computed tomography (CT) apparatus comprising a first detector with a first pixel size, a first CT scan of an object in a first imaging region of the first detector;
obtain a first CT image with a first resolution by reconstructing the first set of projection data by modifying the first pixel size to a second pixel size, wherein the first CT image is used as an input image for a machine-learning model;
obtain a processed CT image with a resolution higher than the first resolution by applying the machine-learning model for resolution enhancement to the first CT image; and
display the processed CT image or output the processed CT image for analysis, wherein
the machine-learning model is obtained by training using a second CT image generated based on a second set of projection data which is acquired by performing, with a second CT apparatus comprising a second detector with the second pixel size smaller than the first pixel size, a second CT scan of the object in a second imaging region of the second detector, the second imaging region being smaller than the first imaging region, wherein the second CT image is used as a target image for the input image.

14. An X-ray computed tomography apparatus, comprising the medical image processing apparatus of claim 13 .

15. A computer readable, non-volatile storage medium storing an image processing program which causes, when executed by a computer, the computer to execute:
obtaining a first set of projection data by performing, with a first computed tomography (CT) apparatus comprising a first detector with a first pixel size, a first CT scan of an object in a first imaging region of the first detector;
obtaining a first CT image with a first resolution by reconstructing the first set of projection data by modifying the first pixel size to a second pixel size, wherein the first CT image is used as an input image for a machine-learning model;
obtaining a processed CT image with a resolution higher than the first resolution by applying the machine-learning model for resolution enhancement to the first CT image; and
displaying the processed CT image or outputting the processed CT image for analysis, wherein
the machine-learning model is obtained by training using a second CT image generated based on a second set of projection data which is acquired by performing, with a second CT apparatus comprising a second detector with the second pixel size smaller than the first pixel size, a

Assignees

• Canon Medical Systems Corp

Inventors

• Lee Tzu-Cheng
• Zhou Jian
• Cai Liang
• Yu Zhou
• Matsuura Masakazu
• Nemoto Takuya
• Taguchi Hiroki

Classifications

• G06T12/10 Primary

  Image preprocessing, e.g. calibration, positioning of sources or scatter correction · CPC title

• G06T12/30

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

• G06T12/00

  Tomographic reconstruction from projections · CPC title

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title

• G06T2207/20081

  Training; Learning · CPC title