Information processing apparatus, information processing method, non-transitory computer readable storage medium, and ultrasonic diagnosis apparatus

Patent metadata

Field	Value
Publication number	US-11723628-B2
Application number	US-202117208934-A
Country	US
Kind code	B2
Filing date	Mar 22, 2021
Priority date	Mar 26, 2020
Publication date	Aug 15, 2023
Grant date	Aug 15, 2023

Abstract

An information processing apparatus includes an acquisition unit configured to acquire ultrasonic image data, an inference unit configured to infer a body mark corresponding to the ultrasonic image data acquired by the acquisition unit, and a display control unit configured to display the body mark inferred by the inference unit together with the ultrasonic image data.

First claim

What is claimed is:

1. An information processing apparatus, comprising:
a memory storing a program; and
one or more processors which, by executing the program, function as:
an acquisition unit configured to acquire ultrasonic image data;
an inference unit configured to infer a body mark by inputting the ultrasonic image data acquired by the acquisition unit to a classifier based on machine learning, wherein the classifier based on machine learning is a classifier that has undergone learning processing based on supervisory data in which information indicating a type of the body mark is set as a correct label and ultrasonic image data captured in an image-capturing range corresponding to the correct label is set as a ground truth image; and
a display control unit configured to provide the body mark inferred by the inference unit together with the ultrasonic image data.

2. The information processing apparatus according to claim 1 ,
wherein the inference unit further infers a position of a probe on the body mark from the ultrasonic image data acquired by the acquisition unit, and
wherein the display control unit displays superimposed image data in which a probe mark based on the inferred position of the probe is superimposed on the body mark inferred by the inference unit.

3. The information processing apparatus according to claim 2 , wherein the inference unit configured to infer a position of a probe on the body mark by inputting the ultrasonic image data acquired by the acquisition unit to a classifier based on machine learning, wherein the classifier based on machine learning is a classifier that has undergone learning processing based on supervisory data in which information indicating the type of the body mark and the position of the probe on the body mark is set as a correct label and ultrasonic image data captured in an image-capturing range corresponding to the correct label is set as a ground truth image.

4. The information processing apparatus according to claim 1 , wherein the inference unit performs inference by a classifier based on machine learning.

5. The information processing apparatus according to claim 4 , wherein the classifier based on machine learning is a classifier that has undergone learning processing based on supervisory data in which information indicating a type of the body mark is set as a correct label and ultrasonic image data captured in an image-capturing range corresponding to the correct label is set as a ground truth image.

6. The information processing apparatus according to claim 4 , wherein the inference unit outputs an inference result as a likelihood.

7. The information processing apparatus according to claim 6 , wherein the display control unit displays the superimposed image data and the likelihood in association with each other.

8. The information processing apparatus according to claim 6 , further comprising a determination unit configured to determine whether the result of inference by the inference unit satisfies a predetermined reference,
wherein the display control unit provides display based on a result of determination by the determination unit.

9. The information processing apparatus according to claim 8 , wherein the determination unit determines whether at least one of a likelihood corresponding to a class constituting the inference result and a difference between likelihoods respectively corresponding to a plurality of classes satisfies the predetermined reference.

10. The information processing apparatus according to claim 9 , wherein the display control unit provides display with a different number of pieces of superimposed image data to be displayed based on the result of determination by the determination unit.

11. The information processing apparatus according to claim 4 , wherein the inference unit includes a plurality of classifiers.

12. The information processing apparatus according to claim 11 ,
wherein the acquisition unit further acquires a type of the probe,
wherein the information processing apparatus further comprises a selection unit configured to select a classifier to perform inference from among the plurality of classifiers constituting the inference unit based on the type of the probe acquired by the acquisition unit, and
wherein the inference unit infers the body mark corresponding to the ultrasonic image data from the ultrasonic image data by using the classifier selected by the selection unit.

13. An information processing method, comprising:
acquiring ultrasonic image data;
inferring a body mark by inputting the acquired ultrasonic image data to a classifier based on machine learning, wherein the classifier based on machine learning is a classifier that has undergone learning processing based on supervisory data in which information indicating a type of the body mark is set as a correct label and ultrasonic image data captured in an image-capturing range corresponding to the correct label is set as a ground truth image; and
displaying the inferred body mark together with the ultrasonic image data.

14. The information processing method according to claim 13 ,
wherein a position of a probe on the body mark is further inferred from the acquired ultrasonic image data, and
wherein superimposed image data in which a probe mark based on the inferred position of the probe is superimposed on the inferred body mark.

15. A non-transitory computer readable storage medium storing a program for causing a computer to execute the information processing method according to claim 13 .

16. An ultrasonic diagnosis apparatus, comprising:
a memory storing a program; and
one or more processors which, by executing the program, function as:
an acquisition unit configured to acquire ultrasonic image data and a type of probe;
a selection unit configured to select a classifier to perform inference from among the plurality of classifiers based on the type of the probe acquired by the acquisition unit;
an inference unit configured to infer a body mark corresponding to the ultrasonic image data by the classifier based on machine learning selected by the selection unit;
a display control unit configured to display the body mark inferred by the inference unit together with the ultrasonic image data.

17. The ultrasonic diagnosis apparatus according to claim 16 ,
wherein the inference unit further infers a position of a probe on the body mark from the ultrasonic image data acquired by the acquisition unit, and
wherein the display control unit displays superimposed image data in which the probe mark based on the inferred position of the probe is superimposed on the body mark inferred by the inference unit.

Assignees

• Canon Kk

Inventors

• Sasaki Shoya
• Iizuka Naoya
• Ueda Ryuta
• Hirano Yoshinori

Classifications

• A61B8/468 Primary

  allowing annotation or message recording · CPC title

• A61B8/4245

  involving determining the position of the probe, e.g. with respect to an external reference frame or to the patient · CPC title

• A61B8/463

  characterised by displaying multiple images or images and diagnostic data on one display · CPC title

• G06F18/217

  Validation; Performance evaluation; Active pattern learning techniques · CPC title

• G06F18/2431

  Multiple classes · CPC title