Information processing system, information processing apparatus, information processing method, and non-transitory storage medium

Patent metadata

Field	Value
Publication number	US-12229225-B2
Application number	US-202117493362-A
Country	US
Kind code	B2
Filing date	Oct 4, 2021
Priority date	Oct 6, 2020
Publication date	Feb 18, 2025
Grant date	Feb 18, 2025

Abstract

An information processing apparatus according to an exemplary embodiment of the present disclosure includes a storage unit configured to store a plurality of trained models including a first trained model for classifying medical image data into a class belonging to a first class hierarchy, and a second trained model for classifying the medical image data into a class belonging to a second class hierarchy lower than the first class hierarchy, an acquisition unit configured to acquire information about inference, a selection unit configured to select a plurality of trained models having a hierarchical relationship from among the stored plurality of trained models, based on the acquired information about inference, and a notification unit configured to provide a notification of the selected plurality of trained models.

First claim

What is claimed is:

1. An information processing apparatus comprising:
a storage unit configured to store a plurality of trained models including a first trained model for classifying inference target image data into a class belonging to a first class hierarchy, and a second trained model for classifying inference target image data into a class belonging to a second class hierarchy lower than the first class hierarchy;
an acquisition unit configured to acquire information about inference;
a selection unit configured to select a plurality of trained models having a hierarchical relationship and performing inference, on inference target image data, in order from a trained model corresponding to a higher class hierarchy from among the stored plurality of trained models, based on the acquired information about inference; and
a notification unit configured to provide a notification of the selected plurality of trained models.

2. The information processing apparatus according to claim 1 , wherein the information about inference is information indicating at least one of an inference target, an inference purpose, or a class.

3. The information processing apparatus according to claim 1 , further comprising an inference unit configured to perform inference using the selected plurality of trained models having the hierarchical relationship.

4. The information processing apparatus according to claim 3 , wherein the inference unit performs the inference starting from a trained model for classifying the medical image data into a class belonging to a highest class hierarchy among the selected plurality of trained models having the hierarchical relationship.

5. The information processing apparatus according to claim 1 , wherein the selection unit calculates a degree of conformity with a trained model corresponding to each class hierarchy, based on the information about inference, and selects the plurality of trained models based on the calculated degree of conformity.

6. The information processing apparatus according to claim 5 , wherein the selection unit selects the plurality of trained models from among trained models with the calculated degree of conformity being greater than a predetermined value.

7. The information processing apparatus according to claim 6 , wherein the selection unit selects a trained model corresponding to a lowest class hierarchy and a trained model that corresponds to a class hierarchy higher than the lowest class hierarchy and has the hierarchical relationship with the trained model corresponding to the lowest class hierarchy, from among the trained models with the calculated degree of conformity being greater than the predetermined value.

8. The information processing apparatus according to claim 6 , wherein the selection unit selects a trained model corresponding to a highest class hierarchy and a trained model that corresponds to a class hierarchy lower than the highest class hierarchy and has the hierarchical relationship with the trained model corresponding to the highest class hierarchy, from among the trained models with the calculated degree of conformity being greater than the predetermined value.

9. The information processing apparatus according to claim 6 , wherein the selection unit selects a trained model corresponding to a highest class hierarchy and a trained model corresponding to a lowest class hierarchy from among the trained models with the calculated degree of conformity being greater than the predetermined value, and in a case where there is another class hierarchy between the highest class hierarchy and the lowest class hierarchy, the selection unit further selects a trained model corresponding to the other class hierarchy.

10. The information processing apparatus according to claim 1 , wherein the acquisition unit further acquires selection information about the trained models, and the selection unit selects the plurality of trained models having the hierarchical relationship, based on the selection information.

11. The information processing apparatus according to claim 5 , wherein the selection unit calculates the degree of conformity by using a trained model for classifying the class belonging to the first class hierarchy and the class belonging to the second class hierarchy.

12. An information processing method comprising:
acquiring information about inference;
selecting a plurality of trained models having a hierarchical relationship and performing inference, on inference target image data, in order from a trained model corresponding to a higher class hierarchy based on the acquired information about inference; and
providing a notification of the selected plurality of trained models.

13. A non-transitory computer-readable storage medium storing a program for causing a computer to execute the information processing method according to claim 12 .

14. An information processing system comprising:
an information processing apparatus including a storage unit configured to store a plurality of trained models including a first trained model for classifying inference target image data into a class belonging to a first class hierarchy, and a second trained model for classifying inference target image data into a class belonging to a second class hierarchy lower than the first class hierarchy; and
an information terminal configured to communicate with the information processing apparatus via a network,
wherein the information processing apparatus includes:
an acquisition unit configured to acquire information about inference;
a selection unit configured to select a plurality of trained models having a hierarchical relationship and performing inference, on inference target image data, in order from a trained model corresponding to a higher class hierarchy from among the stored plurality of trained models, based on the acquired information about inference; and
a notification unit configured to notify the information terminal of the selected plurality of trained models.

15. The information processing system according to claim 14 , wherein the information terminal includes:
a display unit configured to display the plurality of trained models that the information terminal is notified of by the notification unit; and
an operation unit configured to receive an operation to select a trained model to be used in performing the inference, based on the plurality of trained models that the information terminal is notified of by the notification unit.

16. The information processing system according to claim 15 , wherein the acquisition unit in the information processing apparatus acquires selection information about the trained model selected on the operation unit, and the selection unit selects the plurality of trained models having the hierarchical relationship, based on the selection information.

17. The information processing apparatus according to claim 3 , wherein the inference unit performs first inference using the first trained model corresponding to the first class hierarchy among the stored plurality of trained models having the hierarchical relationship and performs second inference using a trained model corresponding to a class hierarchy lower than the first class hierarchy among the stored plurality of trained models having the hierarchical relationship.

18. The information processing apparatus according to claim 17 , wherein based on a result of the first inference, the inference unit performs the second inference.

19. The information processing apparatus according to claim 18 , wherein the inference un

Assignees

• Canon Kk

Inventors

• Umezawa Kohtaro
• Ueda Ryuta

Classifications

• G06F18/285

  Selection of pattern recognition techniques, e.g. of classifiers in a multi-classifier system · CPC title

• G06F18/40

  Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor · CPC title

• G06V2201/03

  Recognition of patterns in medical or anatomical images · CPC title

• G16H30/20 Primary

  for handling medical images, e.g. DICOM, HL7 or PACS · CPC title

• A61B5/7267

  involving training the classification device · CPC title