# US 10,482,604 B2 · Systems and methods for image processing

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

- Number: US 10,482,604 B2
- Document ID: US10482604B2
- Status: Granted
- Kind: B2
- Application number: 15/710,815
- Filed: 2017-09-20
- Published: 2018-11-08
- Granted: 2019-11-19
- Earliest claimed: 2017-05-05
- Assignees: Shanghai United Imaging Healthcare Co., Ltd.
- Applicants: Shanghai United Imaging Healthcare Co., Ltd.
- Inventors: Yanen Guo; Jianhua Shen; Xiaodong Wang
- Web page: https://patentgrep.com/patent/US10482604B2

## CPC

- G06T 7/12: Edge-based segmentation
- G06F 18/2415: based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06T 7/11: Region-based segmentation
- G06T 7/143: involving probabilistic approaches, e.g. Markov random field [MRF] modelling
- G06T 7/149: involving deformable models, e.g. active contour models
- G06T 2207/10072: Tomographic images
- G06T 2207/20076: Probabilistic image processing
- G06T 2207/20081: Training; Learning
- G06T 2207/20116: Active contour; Active surface; Snakes
- G06T 2207/20121: Active appearance model [AAM]
- G06T 2207/20124: Active shape model [ASM]
- G06T 2207/30048: Heart; Cardiac
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A method for image segmentation may include acquiring an image including a region of interest (ROI). The ROI has a first margin, the ROI includes a subregion, and the subregion has a second margin. The method may further include acquiring a first model according to the ROI, wherein the first model has a third margin. The method may further determine, based on the first margin and the third margin, a second model by matching the first model with the image, wherein the second model includes a sub-model, and the sub-model has a fourth margin. The method may further include determining, based on the second margin, a third model by adjusting the fourth margin of the sub-model in the second model. The method may further include segmenting the ROI according to the third model and generating a segmented ROI based on a result of the segmentation.

## Claims

### Claim 1

An image processing method implemented on at least one machine each of which has at least one processor and storage, the method comprising:
acquiring, by the at least one processor, an image including a region of interest (ROI), the ROI having a first margin, the ROI including a subregion, the subregion having a second margin;
acquiring, by the at least one processor, a first model according to the ROI, the first model having a third margin;
determining by the at least one processor, based on the first margin and the third margin; a second model by matching the first model with the image, the second model including a sub-model, the sub-model having a fourth margin;
determining by the at least one processor, based on the second margin, a third model by adjusting the fourth margin of the sub-model of the second model;
segmenting, by the at least one processor, the ROI according to the third model; and
generating, by the at least one processor, a segmented ROI based on a result of the segmentation.

### Claim 2 (depends on 1)

The method of claim 1, wherein the acquiring the first model includes:
acquiring at least one preliminary model corresponding to the ROI;
determining one or more control points of the at least one preliminary model; and
generating, based on the one or more control points, the first model.

### Claim 3 (depends on 2)

The method of claim 2, wherein the generating, based on the one or more control points, the first model includes:
determining a correlation factor based on a relationship between the one or more controls points and a fifth margin of the at least one preliminary model; and
generating, based on the correlation factor and the one or more control points, the first model.

### Claim 4 (depends on 1)

The method of claim 1, wherein the determining the third model includes:
determining a margin point on the fourth margin;
determining, based on the second margin, a target point corresponding to the margin point, the target point being within a range of the fourth margin; and
adjusting the fourth margin based on the margin point and the target point.

### Claim 5 (depends on 4)

The method of claim 4, wherein the determining the target point includes:
determining, based on the margin point, a candidate point within the range of the fourth margin;
determining, based on the second margin, a probability that the candidate point is the target point; and
determining the target point based on the probability.

### Claim 6 (depends on 5)

The method of claim 5, wherein the adjusting the fourth margin of the sub-model includes:
determining, based on the target point, a transformed margin point by performing a similarity transformation on a margin point of the fourth margin; and
adjusting the fourth margin of the sub-model by performing a piecewise-affine transformation on the transformed margin point based on a correlation factor, or by adjusting the transformed margin point based on an energy function.

### Claim 7 (depends on 5)

The method of claim 5, wherein the determining the probability that the candidate point is the target point includes:
obtaining a classifier; and
determining the probability of the candidate point by the classifier.

### Claim 8 (depends on 7)

The method of claim 7, wherein the classifier is generated according to a process for generating a classifier, the process comprising:
obtaining a preliminary classifier;
acquiring a plurality of sample points related to the ROI;
classifying the plurality of sample points into two or more groups; and
training the preliminary classifier to generate the classifier based on the sample points and the two or more groups.

### Claim 9 (depends on 8)

The method of claim 8, wherein the classifying the plurality of sample points into the two or more groups includes:
classifying the plurality of sample points into the two or more groups based on sharpness information or position information of the plurality of sample points.

### Claim 10 (depends on 8)

The method of claim 8, wherein
at least one group of the two or more groups of the sample points include a plurality of positive sample points and a plurality of negative sample points;
the positive sample points reside on or within a range of a sixth margin of the ROI; and
the negative sample points reside beyond a distance from the sixth margin of the ROI.

### Claim 11 (depends on 1)

The method of claim 1, wherein the determining the second model by matching the first model with the image comprises:
determining a margin probability map of the image; and
matching the first model with the margin probability map.

### Claim 12 (depends on 11)

The method of claim 11, wherein the matching the first model with the margin probability map is performed based on a Hough transformation.

### Claim 13

An image processing system, comprising:
at least one processor, and
a storage configured to store instructions, the instructions, when executed by the at least one processor, causing the system to effectuate a method, the method comprising:
acquiring an image including a region of interest (ROI), the ROI having a first margin, the ROI including a subregion, the subregion having a second margin;
acquiring a first model according to the ROI, the first model having a third margin;
determining, based on the first margin and the third margin, a second model by matching the first model with the image, the second model including a sub-model, the sub-model having a fourth margin;
determining, based on the second margin, a third model by adjusting the fourth margin of the sub-model of the second model;
segmenting the ROI according to the third model; and
generating a segmented ROI based on a result of the segmentation.

### Claim 14 (depends on 13)

The system of claim 13, wherein the acquiring the first model includes:
acquiring at least one preliminary model corresponding to the ROI;
determining one or more control points of the at least one preliminary model; and
generating, based on the one or more control points, the first model.

### Claim 15 (depends on 14)

The system of claim 14, wherein the generating, based on the one or more control points, the first model includes:
determining a correlation factor based on a relationship between the one or more controls points and a fifth margin of the at least one preliminary model; and
generating, based on the correlation factor and the one or more control points; the first model.

### Claim 16 (depends on 13)

The system of claim 13, wherein the determining the third model includes:
determining a margin point on the fourth margin;
determining, based on the second margin, a target point corresponding to the margin point, the target point being within a range of the fourth margin; and
adjusting the fourth margin based on the margin point and the target point.

### Claim 17 (depends on 16)

The system of claim 16, wherein the determining the target point includes:
determining, based on the margin point, a candidate point within the range of the fourth margin;
determining, based on the second margin, a probability that the candidate point is the target point; and
determining the target point based on the probability.

### Claim 18 (depends on 17)

The system of claim 17, wherein the determining the probability that the candidate point is the target point includes:
obtaining a classifier; and
determining the probability of the candidate point by the classifier.

### Claim 19 (depends on 18)

The system of claim 18, wherein the classifier is generated according to a process for generating a classifier, the process including:
obtaining a preliminary classifier;
acquiring a plurality of sample points related to the ROI;
classifying the plurality of sample points into two or more groups; and
training the preliminary classifier to generate the classifier based on the sample points and the two or more groups.

### Claim 20

A non-transitory computer readable medium storing instructions, the instructions, when executed by at least one processor; causing the at least one processor to implement a method comprising:
acquiring an image including a region of interest ROI), the ROI having a first margin; the ROI including a subregion, the subregion having a second margin;
acquiring a first model according to the ROI, the first model having a third margin;
determining, based on the first margin and the third margin, a second model by matching the first model with the image, the second model including a sub-model, the sub-model having a fourth margin;
determining, based on the second margin, a third model by adjusting the fourth margin of the sub-model of the second model;
segmenting the ROI according to the third model; and
generating a segmented ROI based on a result of the segmentation.