Medical image processing apparatus, medical image processing method and computer-readable medium

Patent metadata

Field	Value
Publication number	US-12040079-B2
Application number	US-202017120718-A
Country	US
Kind code	B2
Filing date	Dec 14, 2020
Priority date	Jun 15, 2018
Publication date	Jul 16, 2024
Grant date	Jul 16, 2024

Abstract

A medical image processing apparatus includes: an obtaining unit configured to obtain a first image that is a medical image of a predetermined site of a subject; an image quality improving unit configured to generate, from the first image, a second image in which the image quality is improved compared to the first image by using an image quality improving engine that includes a machine learning engine; and a display controlling unit configured to cause a display unit to display a composite image obtained by combining the first image and the second image according to a ratio obtained using information relating to at least a partial region of the first image.

First claim

What is claimed is:

1. A medical image processing apparatus comprising at least one of (a) one or more processors connected to one or more memories storing a program including instructions executed by the one or more processors and (b) circuitry configured to function as:
an obtaining unit configured to obtain a first image of a predetermined site of an eye to be examined; and
a display controlling unit configured to control a display unit to display one of the first image and a second image generated by inputting the first image into an image quality improving engine as an input image of the image quality improving engine, the image quality improving engine including a machine learning engine obtained using an OCTA image as training data, wherein:
in a case where an imaging system of the first image is an imaging system according to OCTA, the second image is generated and the display unit is controlled to display the second image in response to an instruction from an operator; and
in a case where the imaging system of the first image is not the imaging system according to OCTA, the second image is not generated and the display unit is controlled to display the first image.

2. The medical image processing apparatus according to claim 1 , wherein:
the image quality improving engine includes a machine learning engine for which a two-dimensional OCTA image obtained by performing averaging processing is adopted as training data.

3. The medical image processing apparatus according to claim 1 , wherein:
the image quality improving engine includes a machine learning engine for which an image imaged with an imaging apparatus with higher performance than an imaging apparatus used for imaging of the first image, or an image obtained by an imaging process including a number of steps that is greater than a number of steps of an imaging step for imaging the first image is adopted as training data.

4. The medical image processing apparatus according to claim 1 , wherein the at least one of (a) and (b) is further configured to function as:
a determining unit configured to determine whether or not the second image can be generated using the image quality improving engine with respect to the first image.

5. The medical image processing apparatus according to claim 4 , wherein:
the determining unit performs the determination based on at least one of an imaged site, an imaging system, an imaging angle of view and an image size of the first image.

6. The medical image processing apparatus according to claim 4 , wherein:
the second image is generated in response to an instruction from the operator with respect to a determination result obtained by the determining unit.

7. The medical image processing apparatus according to claim 1 , wherein:
the image quality improving engine includes a plurality of image quality improving engines which performed learning using different training data from each other.

8. The medical image processing apparatus according to claim 7 , wherein:
each of the plurality of image quality improving engines performed learning using different training data from each other with respect to at least one of an imaged site, an imaging angle of view, and an image resolution; and
the second image is generated using the image quality improving engine that corresponds to at least one of an imaged site, an imaging angle of view, and an image resolution of the first image.

9. The medical image processing apparatus according to claim 7 , wherein the at least one of (a) and (b) is further configured to function as:
an evaluating unit configured to evaluate, using a machine learning engine that used an evaluation value obtained by a predetermined evaluation technique as training data, image quality of the second image, and
wherein:
a plurality of the second images are generated from the first image using the plurality of image quality improving engines, and
the medical image processing apparatus outputs at least one image among the plurality of second images in accordance with an evaluation result obtained by the evaluating unit.

10. A medical image processing apparatus comprising at least one of (a) one or more processors connected to one or more memories storing a program including instructions executed by the one or more processors and (b) circuitry configured to function as:
an obtaining unit configured to obtain a first image that is a medical image of a predetermined site of a subject;
a generating unit configured to generate a second image with higher image quality than the first image by inputting the first image into an image quality improving engine that includes a machine learning engine; and
an evaluating unit configured to evaluate image quality of the second image,
wherein:
the evaluating unit includes an authenticity evaluating engine that evaluates the authenticity of an image, and
the medical image processing apparatus outputs the second image in a case where an output from the authenticity evaluating engine of the evaluating unit is “True”.

11. The medical image processing apparatus according to claim 10 , wherein:
the authenticity evaluating engine includes a machine learning engine that, as training data, uses an image generated by a different image quality improving engine in which the accuracy of image quality improving processing is lower than the image quality improving engine.

12. The medical image processing apparatus according to claim 1 , wherein the at least one of (a) and (b) is further configured to function as:
an estimating unit configured to estimate, using a machine learning engine which used an image to which a label of at least one of an imaged site and an imaged region is attached as training data, at least one of an imaged site and an imaged region of the first image.

13. The medical image processing apparatus according to claim 1 , wherein:
an image size of the first image is adjusted to an image size which the image quality improving engine is capable of handling and inputs the first image for which the image size was adjusted to the image quality improving engine, and
the second image is generated by adjusting a size of an output image from the image quality improving engine to an original image size of the first image.

14. The medical image processing apparatus according to claim 1 , wherein:
an image size of the first image is adjusted so that a resolution of the first image becomes a predetermined resolution,
with respect to the first image for which the image size was adjusted, an image obtained by performing padding so that the image size which was adjusted becomes an image size which the image quality improving engine is capable of handling is input to the image quality improving engine,
with respect to an output image from the image quality improving engine, trimming is performed so as to trim only a region corresponding to a region in which padding was performed, and
an image size of the image on which trimming was performed is adjusted to an original image size of the first image to generate the second image.

15. The medical image processing apparatus according to claim 1 , wherein:
the first image is divided into a plurality of third images of a predetermined image size,
the plurality of third images are input to the image quality improving engine to generate a plurality of fourth images, and
the plurality of fourth images are integrated to generate the second image.

16. The medical image processing apparatus according to claim 1 , wherein:
the first image with three dimensions is divided into a plurality of

Assignees

• Canon Kk

Inventors

• Mizobe Hideaki
• Iwase Yoshihiko
• Yamazoe Manabu
• Uchida Hiroki
• Tomita Ritsuya

Classifications

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/094

  Adversarial learning · CPC title

• G06N3/0475

  Generative networks · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/096

  Transfer learning · CPC title