# US 11,908,137 B2 · Method, device and equipment for identifying and detecting macular region in fundus image

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

- Number: US 11,908,137 B2
- Document ID: US11908137B2
- Status: Granted
- Kind: B2
- Application number: 18/010,804
- Filed: 2021-08-17
- Published: 2023-08-31
- Granted: 2024-02-20
- Earliest claimed: 2021-04-30
- Assignees: Beijing Zhenhealth Technology Co., Ltd.
- Applicants: Beijing Zhenhealth Technology Co., Ltd.
- Inventors: Dongdong Zhang
- Web page: https://patentgrep.com/patent/US11908137B2

## CPC

- G06T 7/0012: {Biomedical image inspection}
- G06V 40/193: {Preprocessing; Feature extraction}
- G06T 5/40: using histogram techniques
- G06T 7/90: Determination of colour characteristics
- G06V 10/22: by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
- G06V 10/82: using neural networks
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T 2207/30041: Eye; Retina; Ophthalmic
- G06V 2201/03: Recognition of patterns in medical or anatomical images
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

The present disclosure provides a method, device and equipment for identifying and detecting a macular region in a fundus image. The method includes the following steps: reading a current fundus image to be positioned and detected; detecting a macular region in the fundus image using a target detection model; when the macular region in the fundus image is not detected, detecting an optic disk region in the fundus image, and identifying and positioning the macular region based on the detected optic disk region; based on a positioning result of the macular region, extracting a macular image corresponding to the macular region from the fundus image; and performing multi-modal processing on the macular image, fusing images obtained by the multi-modal processing to obtain a fused image, and detecting whether the macular region is qualified or not according to the fused image.

## Claims

### Claim 1

A method for identifying and detecting a macular region in a fundus image, comprising:
reading a current fundus image to be positioned and detected;
detecting a macular region in the fundus image using a target detection model;
when the macular region in the fundus image is not detected, detecting an optic disk region in the fundus image, and identifying and positioning the macular region based on the detected optic disk region;
based on an identification and positioning result of the macular region, extracting a macular image corresponding to the macular region from the fundus image; and
performing multi-modal processing on the macular image, and fusing images obtained by the multi-modal processing to obtain a fused image, wherein
the process of identifying and positioning the macular region based on the detected optic disk region comprises:
obtaining a central coordinate of the optic disk region; and
obtaining a central coordinate of the macular region according to the central coordinate of the optic disk region in combination with a multiple linear regression model, wherein when the central coordinate of the macular region is obtained according to the central coordinate of the optic disk region in combination with the multiple linear regression model, based on a formula:
Y=WX+B,
the central coordinate of the macular region is calculated,
wherein Y is the central coordinate of the macular region, X is the central coordinate of the optic disk region, and both W and B are matrix parameters;
W and B are determined by a least square method;
and the following matrix solving formula is used when W and B are determined by the least square method:
[
w
,
B
]
=
(
X
0
T
⁢
X
0
)
-
1
⁢
X
0
T
⁢
Y
1
⁢
X
0
=
[
X
1
]
1
,
wherein X0 is a matrix vector, consisting of a matrix vector X and a vector 1; and
the images obtained by the multi-modal processing are fused by image channel superposition to obtain the fused image.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the process of performing multi-modal processing on the macular image comprises: performing at least one of contrast limited adaptive histogram equalization (CLARE) processing, Gaussian filter enhancement processing, and hue saturation value (HSV) color space processing on the macular image.

### Claim 3

A device for identifying and detecting a macular region in a fundus image, comprising an image reading module, a first macular region positioning module, a second macular region positioning module, a macular region extraction module, and a macular region detection module, wherein
the image reading module is configured to read a current fundus image to be positioned and detected;
the first macular region positioning module is configured to detect a macular region in the fundus image using a target detection model;
the second macular region positioning module is configured to detect an optic disk region in the fundus image when the macular region in the fundus image is not detected by the first macular region positioning module, and identify and position the macular region based on the detected optic disk region;
the macular region extraction module is configured to extract a macular image corresponding to the macular region from the fundus image based on a positioning result of the macular region; and
the macular region detection module is configured to perform multi-modal processing on the macular image, fuse images obtained by the multi-modal processing to obtain a fused image, and detect whether the macular region is qualified or not according to the fused image, wherein the second macular region positioning module comprises: a coordinate obtaining sub-module and a macular positioning sub-module;
the coordinate obtaining sub-module is configured to obtain a central coordinate of the optic disk region; and
the macular positioning sub-module is configured to obtain a central coordinate of the macular region according to the central coordinate of the optic disk region in combination with a multiple linear regression model, wherein when the macular positioning sub-module is configured to obtain the central coordinate of the macular region according to the central coordinate of the optic disk region in combination with the multiple linear regression model, based on a formula:
Y=WX+B,
the central coordinate of the macular region is calculated,
wherein Y is the central coordinate of the macular region, X is the central coordinate of the optic disk region, and both W and B are matrix parameters;
W and B are determined by a least square method;
and the following matrix solving formula is used when W and B are determined by the least square method:
[
w
,
B
]
=
(
X
0
T
⁢
X
0
)
-
1
⁢
X
0
T
⁢
Y
1
⁢
X
0
=
[
X
1
]
1
,
wherein X0 is a matrix vector, consisting of a matrix vector X and a vector 1; and
the macular region detection module is configured to fuse the images obtained by the multi-modal processing by image channel superposition to obtain the fused image.

### Claim 4 (depends on 3)

The device according to claim 3, wherein the macular region detection module comprises a multi-modal amplification sub-module;
and the multi-modal amplification sub-module is configured to perform multi-modal processing on the macular image,
wherein the process of performing multi-modal processing on the macular image comprises: performing at least one of CLAHE processing, Gaussian filter enhancement processing, and HSV color space processing on the macular image.

### Claim 5 (depends on 1)

Equipment for identifying and detecting a macular region in a fundus image, comprising:
a processor; and
a memory configured to store an instruction executable by the processor,
wherein the processor is configured to implement the method according to claim 1 when executing the executable instruction.

### Claim 6 (depends on 5)

The equipment for identifying and detecting a macular region in a fundus image according to claim 5, wherein the process of performing multi-modal processing on the macular image comprises: performing at least one of contrast limited adaptive histogram equalization (CLAHE) processing, Gaussian filter enhancement processing, and hue saturation value (HSV) color space processing on the macular image.