# US 2021/0241461 A1 · Computer-readable recoding medium having stored therein estimation processing program, estimation processing method and information processing apparatus

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

- Number: US 2021/0241461 A1
- Document ID: US20210241461A1
- Status: Published application
- Kind: A1
- Application number: 17/144,180
- Filed: 2021-01-08
- Published: 2021-08-05
- Earliest claimed: 2020-01-30
- Assignees: Fujitsu Limited; Riken; National Cancer Center
- Applicants: Fujitsu Limited; Riken; National Cancer Center
- Inventors: Akira SAKAI; Masaaki KOMATSU; Kanto SHOZU
- Web page: https://patentgrep.com/patent/US20210241461A1

## CPC

- G06T 7/11: Region-based segmentation
- G06N 5/04: Inference or reasoning models
- G06N 20/00: Machine learning
- G06T 7/0012: {Biomedical image inspection}
- G06T 2207/10132: Ultrasound image
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T 2207/30008: Bone
- G06T 2207/30044: Fetus; Embryo
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A method includes: inputting an image of a target to each of a first model for estimating a region of an annular face of the target, a second model for estimating a region in an outer circumference of the annular face and a third model for estimating a region in an inner circumference of the annular face, and outputting a region image, as an estimated region of the annular face, the region image being obtained by synthesizing first, second, and third elements. The first element is a region outputted as a result of the first model for calculating a sum set in the synthesizing, the second element is a region outputted as a result of the second model for calculating a sum set in the synthesizing, and the third element is a region outputted as a result of the third model for calculating a difference set in the synthesizing.

## Claims

### Claim 1

A non-transitory computer-readable recording medium having stored therein an estimation processing program that causes a computer to execute a process comprising:
inputting an input-image obtained by photographing a target to each of a first estimation model for estimating a region of an annular face of the target, a second estimation model for estimating a region in an outer circumference of the annular face and a third estimation model for estimating a region in an inner circumference of the annular face; and
outputting a region image, as an estimated region of the annular face of the target in the input-image, in response to inputting of the input-image and being used for calculation of a difference set, the region image being obtained by synthesizing a first element, a second element and a third element, the first element being a first region outputted as an estimation result of the first estimation model in response to inputting of the input-image and being used for calculation of a sum set in the synthesizing, the second element being a second region outputted as an estimation result of the second estimation model in response to inputting of the input-image and being used for calculation of a sum set in the synthesizing, the third element being a third region outputted as an estimation result of the third estimation model in response to inputting of the input-image and being used for calculation of a difference set in the synthesizing.

### Claim 2 (depends on 1)

The non-transitory computer-readable recording medium according to claim 1, wherein
the outputting outputs a region obtained by excluding the third region from a region of a sum set of the first region and the second regions as the region estimated as the annular face of the target in the input-image.

### Claim 3 (depends on 1)

The non-transitory computer-readable recording medium according to claim 1, wherein
the outputting outputs a region of a sum set of the first region and a region obtained by excluding the third region from the second region as the region estimated as the annular face of the target in the input-image.

### Claim 4 (depends on 1)

The non-transitory computer-readable recording medium according to claim 1, wherein the process further comprises:
acquiring a dataset of training data each including a training image obtained by photographing a training target and annotation data indicative of a region corresponding to the training target in the training image;
executing machine learning of the first estimation model based on the estimation result acquired in response to inputting of the training image and outputted by the first estimation model and the annotation data as first annotation data;
executing machine learning of the second estimation model based on the estimation result acquired in response to inputting of the training image and outputted by the second estimation model and second annotation data obtained by processing the annotation data so as to indicate a region inside the outer circumference of the annular face of the training target in the training image; and
executing machine learning of the third estimation model based on the estimation result acquired in response to inputting of the training image and outputted by the third estimation model and third annotation data obtained by processing the annotation data so as to indicate a region inside the inner circumference of the annular face of the training target in the training image.

### Claim 5 (depends on 1)

The non-transitory computer-readable recording medium according to claim 1, wherein the process further comprises:
outputting a result of a noise removing process performed for the region estimated as the annular face of the target in the input-image.

### Claim 6 (depends on 1)

The non-transitory computer-readable recording medium according to claim 1, wherein
the input-image is an ultrasonographic image.

### Claim 7

An estimation processing method executed by a computer, the estimation processing method comprising:
inputting an input-image obtained by photographing a target to each of a first estimation model for estimating a region of an annular face of the target, a second estimation model for estimating a region in an outer circumference of the annular face and a third estimation model for estimating a region in an inner circumference of the annular face; and
outputting a region image, as an estimated region of the annular face of the target in the input-image, in response to inputting of the input-image and being used for calculation of a difference set, the region image being obtained by synthesizing a first element, a second element and a third element, the first element being a first region outputted as an estimation result of the first estimation model in response to inputting of the input-image and being used for calculation of a sum set in the synthesizing, the second element being a second region outputted as an estimation result of the second estimation model in response to inputting of the input-image and being used for calculation of a sum set in the synthesizing, the third element being a third region outputted as an estimation result of the third estimation model in response to inputting of the input-image and being used for calculation of a difference set in the synthesizing.

### Claim 8 (depends on 7)

The estimation processing method according to claim 7, wherein
the outputting outputs a region obtained by excluding the third region from a region of a sum set of the first and second regions as the region estimated as the annular face of the target in the input-image.

### Claim 9 (depends on 7)

The estimation processing method according to claim 7, wherein
the outputting outputs a region of a sum set of the first region and a region obtained by excluding the third region from the second region as the region estimated as the annular face of the target in the input-image.

### Claim 10 (depends on 7)

The estimation processing method according to claim 7, further comprising:
acquiring a dataset of training data each including a training image obtained by photographing a training target and annotation data indicative of a region corresponding to the training target in the training image;
executing machine learning of the first estimation model based on the estimation result acquired in response to inputting of the training image and outputted by the first estimation model and the annotation data as first annotation data;
executing machine learning of the second estimation model based on the estimation result acquired in response to inputting of the training image and outputted by the second estimation model and second annotation data obtained by processing the annotation data so as to indicate a region inside the outer circumference of the annular face of the training target in the training image; and
executing machine learning of the third estimation model based on the estimation result acquired in response to inputting of the training image and outputted by the third estimation model and third annotation data obtained by processing the annotation data so as to indicate a region inside the inner circumference of the annular face of the training target in the training image.

### Claim 11 (depends on 7)

The estimation processing method according to claim 7, further comprising:
outputting a result of a noise removing process performed for the region estimated as the annular face of the target in the input-image.

### Claim 12 (depends on 7)

The estimation processing method according to claim 7, wherein
the input-image is an ultrasonographic image.

### Claim 13

An information processing apparatus, comprising:
a memory; and
a processor coupled to the memory, the processor being configured to:
input an input-image obtained by photographing a target to each of a first estimation model for estimating a region of an annular face of the target, a second estimation model for estimating a region in an outer circumference of the annular face and a third estimation model for estimating a region in an inner circumference of the annular face; and
output a region image, as an estimated region of the annular face of the target in the input-image, in response to inputting of the input-image and being used for calculation of a difference set, the region image being obtained by synthesizing a first element, a second element and a third element, the first element being a first region outputted as an estimation result of the first estimation model in response to inputting of the input-image and being used for calculation of a sum set in the synthesizing, the second element being a second region outputted as an estimation result of the second estimation model in response to inputting of the input-image and being used for calculation of a sum set in the synthesizing, the third element being a third region outputted as an estimation result of the third estimation model in response to inputting of the input-image and being used for calculation of a difference set in the synthesizing.

### Claim 14 (depends on 13)

The information processing apparatus according to claim 13, wherein
the processor outputs a region obtained by excluding the third region from a region of a sum set of the first region and the second regions as the region estimated as the annular face of the target in the input-image.

### Claim 15 (depends on 13)

The information processing apparatus according to claim 13, wherein
the processor outputs a region of a sum set of the first region and a region obtained by excluding the third region from the second region as the region estimated as the annular face of the target in the input-image.

### Claim 16 (depends on 13)

The information processing apparatus according to claim 13, wherein the processor:
acquires a dataset of training data each including a training image obtained by photographing a training target and annotation data indicative of a region corresponding to the training target in the training image;
executes machine learning of the first estimation model based on the estimation result acquired in response to inputting of the training image and outputted by the first estimation model and the annotation data as first annotation data;
executes machine learning of the second estimation model based on the estimation result acquired in response to inputting of the training image and outputted by the second estimation model and second annotation data obtained by processing the annotation data so as to indicate a region inside the outer circumference of the annular face of the training target in the training image; and
executes machine learning of the third estimation model based on the estimation result acquired in response to inputting of the training image and outputted by the third estimation model and third annotation data obtained by processing the annotation data so as to indicate a region inside the inner circumference of the annular face of the training target in the training image.

### Claim 17 (depends on 13)

The information processing apparatus according to claim 13, wherein the processor
outputs a result of a noise removing process performed for the region estimated as the annular face of the target in the input-image.

### Claim 18 (depends on 13)

The estimation processing method according to claim 13, wherein
the input-image is an ultrasonographic image.
