# US 12,004,901 B2 · Identifying an optimal image from a number of ultrasound images

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

- Number: US 12,004,901 B2
- Document ID: US12004901B2
- Status: Granted
- Kind: B2
- Application number: 17/257,656
- Filed: 2019-07-05
- Published: 2021-09-02
- Granted: 2024-06-11
- Earliest claimed: 2018-07-06
- Assignees: Koninklijke Philips N.V.
- Applicants: Koninklijke Philips N.V.
- Inventors: Shrutin Ulman; Navaneetha Krishnan Santhana Naidu; Girish Krishnamurthy
- Web page: https://patentgrep.com/patent/US12004901B2

## CPC

- A61B 8/0866: {involving foetal diagnosis; pre-natal or peri-natal diagnosis of the baby}
- A61B 8/5223: {for extracting a diagnostic or physiological parameter from medical diagnostic data (for algorithms to analyse biomedical images G06T7/0012)}
- A61B 8/4236: {characterised by adhesive patches}
- A61B 8/5246: {combining images from the same or different imaging techniques, e.g. color Doppler and B-mode}
- A61B 8/5292: {using additional data, e.g. patient information, image labeling, acquisition parameters}
- G06T 5/40: using histogram techniques
- G06T 7/0014: {using an image reference approach}
- G16H 30/20: for handling medical images, e.g. DICOM, HL7 or PACS
- G16H 50/30: for calculating health indices; for individual health risk assessment
- G06T 2207/10132: Ultrasound image
- G06T 2207/30044: Fetus; Embryo
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA

## Abstract

The invention relates to a method of identifying, from a number of ultrasound images (34) acquired during an ultrasound examination of a subject (6), the ultrasound image best suited to analyze a pre-determined anatomical structure of the subject. The method comprises the steps of (a) Providing a number of ultrasound images (34) acquired during an ultrasound examination of the subject; (b) Deriving at least one biometric parameter (96) related to the anatomical structure from a physical examination of the subject, (c) Retrieving a reference image (38) showing a target view of the anatomical structure (35); (d) For each of the number of ultrasound images (34), calculating an image similarity index between the ultrasound image and the reference image, (e) For at least the ultrasound image(s) having the best image similarity index, determining (100) a biometric parameter from the ultrasound image, and calculating a biometric similarity index, i.e. a measure of the agreement between the biometric parameter, as determined from the ultrasound image, and the corresponding biometric parameter as derived from the physical examination; and (f) selecting the optimal ultrasound image based on the biometric similarity index, and optionally on the image similarity index.

## Claims

### Claim 1

A method of identifying, from a number of ultrasound images acquired during an ultrasound examination of a subject, an ultrasound image best suited to analyze a pre-determined anatomical structure of the subject, the method comprising the steps of:
providing a number of ultrasound images acquired during an ultrasound examination of the subject;
providing at least one biometric parameter related to the anatomical structure, the biometric parameter being derived from a physical examination of the subject, and/or from the subject's medical history;
retrieving a reference image from a database, the reference image showing a target view of the anatomical structure;
for each of the number of ultrasound images, calculating an image similarity index between the ultrasound image and the reference image;
for at least the ultrasound image(s) having an image similarity index above a certain threshold, or having the best image similarity index, determining a biometric parameter from the ultrasound image, and calculating a biometric similarity index, which is a measure of the agreement between the biometric parameter, as determined from the ultrasound image, and the corresponding biometric parameter as derived from the physical examination and/or the subject's medical history;
based on the biometric similarity index, and optionally on the image similarity index, selecting an ultrasound image, the selected ultrasound image, as being best suited to analyze the anatomical structure; and
for at least the selected ultrasound image, automatically fitting a pre-determined curve or geometrical shape to an anatomical structure shown in the selected ultrasound image, and determining at least a coefficient of the fitted geometrical shape or curve.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the reference image is selected from a number of reference images of the anatomical structure, based on at least one biometric parameter derived from the physical examination of the subject, and/or from the subject's medical history.

### Claim 3 (depends on 1)

The method according to claim 1, wherein the subject is a pregnant woman, and the anatomical structure is or includes a fetus.

### Claim 4 (depends on 1)

The method according to claim 1, wherein the image similarity index between an ultrasound image and the reference image is calculated by one or more of the following steps:
identifying and comparing characteristic features, wherein the characteristic feature are at least one of anatomical features or geometric features, of the ultrasound image and the reference image;
computing and comparing grey level histograms of pixel values of the ultrasound image and the reference image.

### Claim 5 (depends on 1)

The method according to claim 1, wherein the biometric similarity index is calculated by automatically or manually taking measurements on the anatomical structure on at least the ultrasound image(s) having an image similarity index above a certain threshold, or having the best image similarity index, and calculating the biometric similarity index therefrom.

### Claim 6 (depends on 1)

The method according to claim 1, further comprising a step of calculating the biometric similarity index from the coefficient of the fitted curve or geometrical shape.

### Claim 7 (depends on 1)

The method according to claim 1, further comprising the steps of:
comparing the coefficient of the fitted curve or geometrical shape to the coefficient of a corresponding geometrical shape or curve fitted to the reference image, and
calculating a geometric similarity index which is a measure of the agreement between the coefficient derived from the reference image and the coefficient derived from the selected ultrasound image, and/or, if the geometrical shape is an axis, calculating an axis similarity index by comparing the length or orientation of said axis to the length and/or orientation of an axis fitted to the anatomical structure in the reference index,
and wherein the selection of an ultrasound image as being best suited to analyze the anatomical structure is also based on the geometric similarity index and/or the axis similarity index.

### Claim 8 (depends on 1)

The method according to claim 1, including, for each selected ultrasound image for which a pre-determined curve or geometrical shape is fitted to an anatomical structure shown in the ultrasound image calculated, the steps of
displaying the fitted curve or geometrical shape or axis overlaid on the selected ultrasound image;
optionally providing an operator with the opportunity to correct the fitted curve or geometrical shape or axis.

### Claim 9 (depends on 7)

The method according to claim 7, wherein a matching index between an ultrasound image and the reference image is calculated for at least one of the number of ultrasound images, wherein the matching index is based on the biometric similarity index and one or more of the image similarity index, the geometric similarity index and the axis similarity index.

### Claim 10 (depends on 9)

The method according to claim 9, wherein the number of ultrasound images corresponds to a series of live images acquired continuously by an operator, and wherein at least one of the matching index, the biometric similarity index, the image similarity index, the geometric similarity index and the axis similarity index is calculated in real time for at least some of the live images directly after their acquisition, and wherein a graphical or phonetic signal corresponding to one or several of the matching index and the similarity indices is issued to the operator for at least some of the live images directly after their acquisition.

### Claim 11

A system configured to assist an operator of an ultrasound system in obtaining an ultrasound image well suited to analyze a pre-determined anatomical structure of a subject, the system comprising a processor configured for:
receiving a number of ultrasound images acquired during an ultrasound examination of the subject;
receiving at least one biometric parameter related to the anatomical structure derived from a physical examination of the subject and/or the subject's medical history;
retrieving a reference image from a database, the reference image showing a target view of the anatomical structure;
for each of the number of ultrasound images, calculating an image similarity index between the ultrasound image and the reference image,
for at least the ultrasound image(s) having an image similarity index above a certain threshold, or having the best image similarity index, determining a biometric parameter from the ultrasound image, and calculating a biometric similarity index, which is a measure of the agreement between the biometric parameter, as determined from the ultrasound image, and the corresponding biometric parameter as derived from the physical examination and/or the subject's medical history;
performing at least one of: displaying a graphical representation of at least one of the biometric similarity index, the similarity index, or a matching index calculated therefrom;
and selecting an ultrasound image, the selected ultrasound image, as being best suited to analyze the anatomical structure based on the biometric similarity index, and optionally on the image similarity index; and
for at least the selected ultrasound image, automatically fitting a pre-determined curve or geometrical shape to an anatomical structure shown in the selected ultrasound image, and determining at least a coefficient of the fitted geometrical shape or curve.

### Claim 12 (depends on 11)

The system of claim 11, further comprising an ultrasound system comprising an ultrasound probe handled by the operator, and configured for acquiring a series of live images from the subject; and
wherein the processor is operatively coupled to the ultrasound system and configured for receiving the number of ultrasound images from the ultrasound system as a series of live images.

### Claim 13 (depends on 12)

The system of claim 12, further comprising an adhesive label for affixing to the body surface of the subject, wherein the adhesive label includes markings indicating positions and/or movement directions and/or angulations of the ultrasound probe, which are configured to give directions to the operator to position the probe in a suitable way in order to obtain an ultrasound image well suited to analyze the anatomical structure.

### Claim 14

A non-transitory computer-readable medium comprising instructions that when executed on a processor, cause the processor to:
provide at least one biometric parameter related to the anatomical structure, the biometric parameter being derived from a physical examination of the subject, and/or from the subject's medical history;
retrieving a reference image from a database, the reference image showing a target view of the anatomical structure;
for each of the number of ultrasound images, calculating an image similarity index between the ultrasound image and the reference image;
for at least the ultrasound image(s) having an image similarity index above a certain threshold, or having the best image similarity index, determining a biometric parameter from the ultrasound image, and calculating a biometric similarity index, which is a measure of the agreement between the biometric parameter, as determined from the ultrasound image, and the corresponding biometric parameter as derived from the physical examination and/or the subject's medical history;
based on the biometric similarity index, and optionally on the image similarity index, selecting an ultrasound image as being best suited to analyze the anatomical structure;
for at least one of the number of ultrasound images, automatically fitting a pre-determined curve or geometrical shape to an anatomical structure shown in the ultrasound image, and determining at least a coefficient of the fitted geometrical shape or curve.