# US 2023/0263391 A1 · Ophthalmic information processing apparatus, ophthalmic apparatus, ophthalmic information processing method, and recording medium

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0263391 A1
- Document ID: US20230263391A1
- Status: Published application
- Kind: A1
- Application number: 18/142,043
- Filed: 2023-05-02
- Published: 2023-08-24
- Earliest claimed: 2020-11-04
- Assignees: Topcon Corporation; Tohoku University; Riken
- Applicants: Topcon Corporation; Tohoku University; Riken
- Inventors: Toru NAKAZAWA; Kazuko OMODAKA; Hideo YOKOTA; Guangzhou AN; Tsutomu KIKAWA
- Web page: https://patentgrep.com/patent/US20230263391A1

## CPC

- A61B 3/12: for looking at the eye fundus, e.g. ophthalmoscopes (A61B3/13 takes precedence)
- A61B 3/14: Arrangements specially adapted for eye photography
- G06T 7/0012: {Biomedical image inspection}
- G06T 2207/10024: Color image
- G06T 2207/10101: Optical tomography; Optical coherence tomography [OCT]
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T 2207/30041: Eye; Retina; Ophthalmic
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

An ophthalmic information processing apparatus includes a determiner and a detector. The determiner is configured to determine a presence or absence of a disc hemorrhage for a front image of a fundus of a subject's eye, using a disc hemorrhage determination model obtained by performing machine learning using a plurality of fundus images labeled with labels indicating the presence or absence of a disc hemorrhage as first teaching data. The detector is configured to detect a disc hemorrhage region depicted in the front image that is determined to have the disc hemorrhage, using a disc hemorrhage region detection model obtained by performing machine learning using a plurality of pairs of image groups as second teaching data, each pair having a front image of a fundus and a disc hemorrhage region image representing a disc hemorrhage region depicted in the front image.

## Claims

### Claim at position 1

1: An ophthalmic information processing apparatus, comprising:
a determiner circuit configured to determine a presence or absence of a disc hemorrhage for a front image of a fundus of a subject's eye, using a disc hemorrhage determination model obtained by performing machine learning using a plurality of fundus images labeled with labels indicating the presence or absence of a disc hemorrhage as first teaching data; and
a detector circuit configured to detect a disc hemorrhage region depicted in the front image that is determined to have the disc hemorrhage by the determiner, using a disc hemorrhage region detection model obtained by performing machine learning using a plurality of pairs of image groups as second teaching data, each pair having a front image of a fundus and a disc hemorrhage region image representing a disc hemorrhage region depicted in the front image.

### Claim at position 2 (depends on 1)

2: The ophthalmic information processing apparatus of claim 1, wherein
the second teaching data includes a front image included in the first teaching data.

### Claim at position 3 (depends on 1)

3: The ophthalmic information processing apparatus of claim 1, further comprising
a first learning circuit configured to generate the disc hemorrhage determination model by performing supervised machine learning using the first teaching data.

### Claim at position 4 (depends on 1)

4: The ophthalmic information processing apparatus of claim 1, further comprising
a second learning circuit configured to generate the disc hemorrhage region detection model by performing supervised machine learning using the second teaching data.

### Claim at position 5 (depends on 1)

5: The ophthalmic information processing apparatus of claim 1, wherein
the front image of the fundus is a color front image.

### Claim at position 6 (depends on 1)

6: The ophthalmic information processing apparatus of claim 1, further comprising
an analyzer circuit configured to generate at least one of position information of a position of the disc hemorrhage region in the front image of the fundus of the subject's eye, shape information representing a shape of the disc hemorrhage region, or occurrence information representing occurrence status of the disc hemorrhage region, by analyzing the front image of the fundus of the subject's eye.

### Claim at position 7 (depends on 6)

7: The ophthalmic information processing apparatus of claim 6, wherein
the position information includes information indicating a direction of a representative position of the disc hemorrhage region relative to a reference position.

### Claim at position 8 (depends on 6)

8: The ophthalmic information processing apparatus of claim 6, wherein
the position information includes information indicating whether the representative position of the disc hemorrhage region is within or outside an optic disc region.

### Claim at position 9 (depends on 8)

9: The ophthalmic information processing apparatus of claim 8, wherein
the position information includes information indicating whether the representative position is within or outside a neuroretinal rim.

### Claim at position 10 (depends on 6)

10: The ophthalmic information processing apparatus of claim 6, wherein
the shape information includes at least one of an ellipticity of the disc hemorrhage region or an area of the disc hemorrhage region.

### Claim at position 11 (depends on 6)

11: The ophthalmic information processing apparatus of claim 6, wherein
the occurrence information includes at least one of an occurrence frequency of the disc hemorrhage region or an occurrence interval of the disc hemorrhage region.

### Claim at position 12 (depends on 6)

12: The ophthalmic information processing apparatus of claim 6, further comprising
a position matching circuit configured to perform position matching between OCT data of the subject's eye and the front image of the subject's eye in which the disc hemorrhage region detected by the detector is depicted, wherein
the analyzer circuit is configured to generate at least one of the position information, the shape information, or the occurrence information, using the OCT data on which position matching has been performed with the front image by the position matching circuit.

### Claim at position 13 (depends on 1)

13: An ophthalmic apparatus, comprising:
an imaging unit including a camera and configured to image the fundus of the subject's eye; and
the ophthalmic information processing apparatus of claim 1.

### Claim at position 14 (depends on 12)

14: An ophthalmic apparatus, comprising:
an imaging unit including a camera and configured to image the fundus of the subject's eye;
an OCT device including a scanner and configured to acquire the OCT data by performing optical coherence tomography on the subject's eye; and
the ophthalmic information processing apparatus of claim 12.

### Claim at position 15

15: An ophthalmic information processing method, comprising:
a determination step of determining a presence or absence of a disc hemorrhage for a front image of a fundus of a subject's eye, using a disc hemorrhage determination model obtained by performing machine learning using a plurality of fundus images labeled with labels indicating the presence or absence of a disc hemorrhage as first teaching data; and
a detection step of detecting a disc hemorrhage region depicted in the front image that is determined to have the disc hemorrhage in the determination step, using a disc hemorrhage region detection model obtained by performing machine learning using a plurality of pairs of image groups as second teaching data, each pair having a front image of a fundus and a disc hemorrhage region image representing a disc hemorrhage region depicted in the front image.

### Claim at position 16 (depends on 15)

16: The ophthalmic information processing method of claim 15, wherein
the second teaching data includes a front image included in the first teaching data.

### Claim at position 17 (depends on 15)

17: The ophthalmic information processing method of claim 15, further comprising
a first learning step of generating the disc hemorrhage determination model by performing supervised machine learning using the first teaching data.

### Claim at position 18 (depends on 15)

18: The ophthalmic information processing method of claim 15, further comprising
a second learning step of generating the disc hemorrhage region detection model by performing supervised machine learning using the second teaching data.

### Claim at position 19 (depends on 15)

19: The ophthalmic information processing method of claim 15, wherein
the front image of the fundus is a color front image.

### Claim at position 20 (depends on 15)

20: The ophthalmic information processing method of claim 15, further comprising
an analyzing step of generating at least one of position information of a position of the disc hemorrhage region in a front image of the fundus of the subject's eye, shape information representing a shape of the disc hemorrhage region, or occurrence information representing occurrence status of the disc hemorrhage region, by analyzing the front image of the fundus of the subject's eye.

### Claim at position 21 (depends on 20)

21: The ophthalmic information processing method of claim 20, wherein
the position information includes information indicating a direction of a representative position of the disc hemorrhage region relative to a reference position.

### Claim at position 22 (depends on 20)

22: The ophthalmic information processing method of claim 20, wherein
the position information includes information indicating whether the representative position of the disc hemorrhage region is within or outside an optic disc region.

### Claim at position 23 (depends on 22)

23: The ophthalmic information processing method of claim 22, wherein
the position information includes information indicating whether the representative position is within or outside a neuroretinal rim.

### Claim at position 24 (depends on 20)

24: The ophthalmic information processing method of claim 20, wherein
the shape information includes at least one of an ellipticity of the disc hemorrhage region or an area of the disc hemorrhage region.

### Claim at position 25 (depends on 20)

25: The ophthalmic information processing method of claim 20, wherein
the occurrence information includes at least one of an occurrence frequency of the disc hemorrhage region or an occurrence interval of the disc hemorrhage region.

### Claim at position 26 (depends on 20)

26: The ophthalmic information processing method of claim 20, further comprising
a position matching step of performing position matching between OCT data of the subject's eye and the front image of the subject's eye in which the disc hemorrhage region detected in the detection step, wherein
the analyzing step is performed to generate at least one of the position information, the shape information, or the occurrence information, using the OCT data on which position matching has been performed with the front image in the position matching step.

### Claim at position 27 (depends on 15)

27: A computer readable non-transitory recording medium in which a program for causing a computer to execute each step of the ophthalmic information processing method of claim 15 is recorded.
