# US 10,769,778 B2 · Discriminating between normal and abnormal left ventricles in echocardiography

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

- Number: US 10,769,778 B2
- Document ID: US10769778B2
- Status: Granted
- Kind: B2
- Application number: 15/222,077
- Filed: 2016-07-28
- Published: 2016-11-17
- Granted: 2020-09-08
- Earliest claimed: 2014-04-27
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: David Beymer; Patrick McNeillie; Tanveer Syeda-Mahmood; Quan Wang
- Web page: https://patentgrep.com/patent/US10769778B2

## CPC

- G06T 7/0014: {using an image reference approach}
- A61B 8/0883: {for diagnosis of the heart}
- A61B 8/5223: {for extracting a diagnostic or physiological parameter from medical diagnostic data (for algorithms to analyse biomedical images G06T7/0012)}
- G06F 18/22: Matching criteria, e.g. proximity measures
- G06F 18/23: Clustering techniques
- G06F 18/24: Classification techniques
- G06T 7/0012: {Biomedical image inspection}
- G06T 7/11: Region-based segmentation
- G06T 7/13: Edge detection
- G06T 7/60: Analysis of geometric attributes
- G06T 7/74: {involving reference images or patches}
- G06V 10/755: {Deformable models or variational models, e.g. snakes or active contours}
- G06V 10/7553: {based on shape, e.g. active shape models [ASM]}
- G06F 18/2411: based on the proximity to a decision surface, e.g. support vector machines
- G06T 2207/10132: Ultrasound image
- G06T 2207/20021: Dividing image into blocks, subimages or windows
- G06T 2207/20112: Image segmentation details
- G06T 2207/30048: Heart; Cardiac
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Embodiments of the present invention relate to discriminating between normal and abnormal left ventricles in echocardiography. In some embodiments, a first region of a first image of a heart is located by clustering. The first region depicts a chamber of the heart. The chamber has a chamber shape. A boundary of the first region is determined by contour tracing a retaining the traced contour overlapping the first region. A predetermined shape is compared to the boundary. The predetermined shape had been determined without reference to the first image. A first plurality of parameters is determined that, when applied to the predetermined shape, conforms the predetermined shape to the boundary and minimizes at least one error between the predetermined shape and the boundary. An indication of normality or abnormality is determine from the first plurality of parameters.

## Claims

### Claim 1

A method comprising:
determining an ultrasound scan sector of an ultrasound image of a heart, the ultrasound scan sector comprising a border defined by a first dominant line, a second dominant line, and a side of the ultrasound image, said first and second dominant lines being within the ultrasound image and intersecting at an intersection point;
partitioning the ultrasound scan sector into a plurality of regions by performing clustering on the ultrasound scan sector;
determining a distance between each partitioned region and the intersection point;
selecting a target region of the plurality of regions, said target region having a smallest distance to the intersection point, the target region depicting a chamber shape of a chamber of the heart;
performing contour tracing on the target region to thereby determine a traced contour and retaining at least a portion of the traced contour overlapping the target region to thereby determine a boundary of the chamber, said boundary depicting the chamber shape;
comparing a predetermined shape to the retained, overlapping boundary, the predetermined shape having been determined without reference to the ultrasound image;
determining a plurality of parameters so that, when the determined plurality of parameters are applied to the predetermined shape, conforms the predetermined shape to the boundary and minimizes at least one error between the predetermined shape and the boundary, the plurality of parameters comprising a first axis, a second axis, and a distortion parameter; and
determining from the plurality of parameters an indication of normality or abnormality of the chamber shape.

### Claim 2 (depends on 1)

The method of claim 1, wherein the ultrasound image includes a four-chamber view.

### Claim 3 (depends on 1)

The method of claim 1, wherein the chamber is a left ventricle.

### Claim 4 (depends on 1)

The method of claim 1, further comprising:
determining a fit between the predetermined shape and the retained, overlapping boundary, and wherein the indication of normality or abnormality is determined from the fit.

### Claim 5 (depends on 1)

The method of claim 1, wherein the predetermined shape is selected from the group consisting of: an ellipsoid, a spheroid and a distorted ellipse.

### Claim 6 (depends on 1)

The method of claim 1, further comprising:
estimating a scale of the ultrasound image.

### Claim 7 (depends on 1)

The method of claim 1, further comprising:
receiving a video; and
extracting the ultrasound image from the video.

### Claim 8 (depends on 7)

The method of claim 7, further comprising:
extracting a plurality of additional ultrasound images from the video.

### Claim 9 (depends on 1)

The method of claim 1, wherein the target region is a first target region, the method further comprising:
selecting a second target region in the ultrasound image of the heart, the second target region comprising the first target region, and the second target region depicting four chambers of the heart.

### Claim 10 (depends on 9)

The method of claim 9, wherein selecting the first target region comprises:
thresholding the second target region; and
determining a location of the first target region within the second target region.

### Claim 11 (depends on 10)

The method of claim 10, wherein determining the location of the first target region within the second target region comprises:
performing k-means clustering.

### Claim 12 (depends on 1)

The method of claim 1, wherein determining the plurality of parameters comprises:
estimating a centroid of the retained, overlapping boundary;
determining a dominant orientation of the ultrasound image, the dominant orientation comprising the first and second dominant lines and a sector angle between the first and second dominant lines; and
determining a rotation of the chamber using the dominant orientation.

### Claim 13 (depends on 1)

The method of claim 1, wherein selecting the target region of the ultrasound image comprises:
determining a dominant orientation within the ultrasound image, the dominant orientation comprising the first and second dominant lines and a sector angle between the first and second dominant lines; and
performing thresholding using the ultrasound image.

### Claim 14 (depends on 1)

The method of claim 1, wherein determining the ultrasound scan sector comprises:
determining first and second high intensity spots by applying a Hough transform to the ultrasound image, wherein the first high intensity spot corresponds to the first dominant line, the second high intensity spot corresponds to the second dominant line, wherein the first and second high intensity spots both correspond to a same radius;
determining a horizontal separation of the first and second high intensity spots, wherein the horizontal separation corresponds to a sector angle between the first and second dominant lines.

### Claim 15

A computer program product for discriminating between normal and abnormal left ventricles in echocardiography, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:
determine an ultrasound scan sector of an ultrasound image of a heart, the ultrasound scan sector comprising a border defined by a first dominant line, a second dominant line, and a side of the ultrasound image, said first and second dominant lines being within the ultrasound image and intersecting at an intersection point;
partition the ultrasound scan sector into a plurality of regions by performing clustering on the ultrasound scan sector;
determine a distance between each partitioned region and the intersection point;
select a target region of the plurality of regions, said target region having a smallest distance to the intersection point, the target region depicting a chamber shape of a chamber of the heart;
perform contour tracing on the target region to thereby determine a traced contour and retaining at least a portion of the traced contour overlapping the target region to thereby determine a boundary of the chamber, said boundary depicting the chamber shape;
compare a predetermined shape to the retained, overlapping boundary, the predetermined shape having been determined without reference to the ultrasound image;
determine a plurality of parameters that, when applied to the predetermined shape, conforms the predetermined shape to the boundary and minimizes at least one error between the predetermined shape and the boundary, the plurality of parameters comprising a first axis, a second axis, and a distortion parameter; and
determine from the plurality of parameters an indication of normality or abnormality of the chamber shape.

### Claim 16 (depends on 15)

The computer program product of claim 15, the program instructions further executable by the processor to cause the processor to:
determine a fit between the predetermined shape and the retained, overlapping boundary, and wherein the indication of normality or abnormality is determined from the fit.

### Claim 17 (depends on 15)

The computer program product of claim 15, the program instructions further executable by the processor to cause the processor to:
receive a video; and
extract the ultrasound image from the video.

### Claim 18 (depends on 15)

The computer program product of claim 15, wherein the target region is a first target region, the program instructions further executable by the processor to cause the processor to:
select a second target region in the ultrasound image of the heart, the second target region comprising the first target region, and the second target region depicting four chambers of the heart;
wherein select the first target region comprises:
threshold the second target region; and
determine a location of the first target region within the second target region.

### Claim 19 (depends on 15)

The computer program product of claim 15, wherein determine the plurality of parameters comprises:
estimate a centroid of the retained, overlapping boundary;
determine a dominant orientation of the ultrasound image, the dominant orientation comprising the first and second dominant lines and a sector angle between the first and second dominant lines; and
determine a rotation of the chamber using the dominant orientation.

### Claim 20 (depends on 15)

The computer program product of claim 15, wherein select the target region of the ultrasound image comprises:
determine a dominant orientation within the ultrasound image, the dominant orientation comprising the first and second dominant lines and a sector angle between the first and second dominant lines; and
perform thresholding using the ultrasound image.