# US 2015/0104087 A1 · Automated Fundus Image Field Detection and Quality Assessment

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

- Number: US 2015/0104087 A1
- Document ID: US20150104087A1
- Status: Published application
- Kind: A1
- Application number: 14/511,936
- Filed: 2014-10-10
- Published: 2015-04-16
- Earliest claimed: 2013-10-10
- Assignees: University of Rochester; Rochester Institute of Technology
- Inventors: Gajendra Jung Katuwal; John P. Kerekes; Rajeev S. Ramchandran; Christye P. Sisson
- Web page: https://patentgrep.com/patent/US20150104087A1

## CPC

- G06T 7/11: Region-based segmentation
- G06T 7/0002: {Inspection of images, e.g. flaw detection}
- G06T 7/194: involving foreground-background segmentation
- G06T 7/30: Determination of transform parameters for the alignment of images, i.e. image registration
- G06T 2207/30041: Eye; Retina; Ophthalmic
- G06T 2207/30101: Blood vessel; Artery; Vein; Vascular
- G06T 2207/30168: Image quality inspection
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A method, system, and computer readable medium which automatically determine the side, field and a level of image quality of fundus images of the retina of a human eye is disclosed. The disclosure combines image processing, computer vision and pattern recognition techniques in a unique way to provide a robust process to identify and grade the quality of fundus images with application to improve efficiency and reduce errors in clinical or diagnostic retinal imaging workflows.

## Claims

### Claim 1

A method for determining the sufficiency of a set of images for use in retinal screening, comprising:
identifying the eye and field of each image from a plurality of retinal images comprising a plurality of different fields;
determining the quality of each identified image; and
assigning a comprehensive image quality metric to a set of images from the plurality of retinal images, wherein the set of images is associated with the same eye.

### Claim 2 (depends on 1)

The method of claim 1, wherein the plurality of retinal images comprises a set of three image fields associated with the left eye and a set of three image fields associated with the right eye.

### Claim 3 (depends on 2)

The method of claim 2, wherein the set of three image fields comprises a disc centered field, macula centered field, and temporal to macula field.

### Claim 4 (depends on 1)

The method of claim 1, wherein the identifying is performed by determining where physiological landmarks are located in each image.

### Claim 5 (depends on 4)

The method of claim 4, wherein the physiological landmarks comprise an optic disc and a macula.

### Claim 6 (depends on 1)

The method of claim 1, wherein the determining is performed by analysis of the vessel symmetry of the eye in the identified image.

### Claim 7 (depends on 6)

The method of claim 6, wherein the analysis of the vessel symmetry is performed by a vessel segmentation method.

### Claim 8 (depends on 1)

The method of claim 1, wherein the assigning is performed by image registration among the set of field images associated with the same eye by analyzing selected overlapping areas of each of the image fields.

### Claim 9 (depends on 8)

The method of claim 8, wherein the comprehensive metric is calculated by assigning a quality metric to each area and selecting the highest quality metric for each overlapping area and averaging the assigned metrics.

### Claim 10

A computer readable medium having stored thereon instructions for determining the sufficiency of a set of images for use in retinal screening, comprising machine executable code which when executed by at least one processor, causes the processor to perform actions comprising:
identifying the eye and field of each image from a plurality of retinal images comprising a plurality of different fields;
determining the quality of each identified image; and
assigning a comprehensive image quality metric to a set of images from the plurality of retinal images, wherein the set of images is associated with the same eye.

### Claim 11 (depends on 10)

The medium of claim 10, wherein the plurality of retinal images comprises a set of three image fields associated with the left eye and a set of three image fields associated with the right eye.

### Claim 12 (depends on 11)

The method of claim 11, wherein the set of three image fields comprises a disc centered field, macula centered field, and temporal to macula field.

### Claim 13 (depends on 10)

The method of claim 10, wherein the identifying is performed by determining where physiological landmarks are located in each image.

### Claim 14 (depends on 13)

The method of claim 13, wherein the physiological landmarks comprise an optic disc and a macula.

### Claim 15 (depends on 10)

The method of claim 10, wherein the determining is performed by analysis of the vessel symmetry of the eye in the identified image.

### Claim 16 (depends on 15)

The method of claim 15, wherein the analysis of the vessel symmetry is performed by a vessel segmentation method.

### Claim 17 (depends on 10)

The method of claim 10, wherein the assigning is performed by image registration among the set of field images associated with the same eye by analyzing selected overlapping areas of each of the image fields.

### Claim 18 (depends on 17)

The method of claim 17, wherein the comprehensive metric is calculated by assigning a quality metric to each area and selecting the highest quality metric for each overlapping area and averaging the assigned metrics.

### Claim 19

A system for determining the sufficiency of a set of images for use in retinal screening, comprising:
a processor which performs the actions comprising:
identifying the eye and field of each image from a plurality of retinal images comprising a plurality of different fields;
determining the quality of each identified image; and
assigning a comprehensive image quality metric to a set of images from the plurality of retinal images, wherein the set of images is associated with the same eye.

### Claim 20 (depends on 19)

The processor of claim 19, wherein the plurality of retinal images comprises a set of three image fields associated with the left eye and a set of three image fields associated with the right eye.

### Claim 21 (depends on 20)

The processor of claim 20, wherein the set of three image fields comprises a disc centered field, macula centered field, and temporal to macula field.

### Claim 22 (depends on 19)

The processor of claim 19, wherein the identifying is performed by determining where physiological landmarks are located in each image.

### Claim 23 (depends on 22)

The processor of claim 22, wherein the physiological landmarks comprise an optic disc and a macula.

### Claim 24 (depends on 19)

The processor of claim 19, wherein the determining is performed by analysis of the vessel symmetry of the eye in the identified image.

### Claim 25 (depends on 24)

The processor of claim 24, wherein the analysis of the vessel symmetry is performed by a vessel segmentation method.

### Claim 26 (depends on 19)

The processor of claim 19, wherein the assigning is performed by image registration among the set of field images associated with the same eye by analyzing selected overlapping areas of each of the image fields.

### Claim 27 (depends on 26)

The processor of claim 26, wherein the comprehensive metric is calculated by assigning a quality metric to each area and selecting the highest quality metric for each overlapping area and averaging the assigned metrics.
