# US 2023/0397898 A1 · Method and system for material decomposition in dual-or multiple-energy x-ray based imaging

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

- Number: US 2023/0397898 A1
- Document ID: US20230397898A1
- Status: Published application
- Kind: A1
- Application number: 18/211,056
- Filed: 2023-06-16
- Published: 2023-12-14
- Earliest claimed: 2022-06-14
- Assignees: Curvebeam Ai Limited
- Applicants: Curvebeam Ai Limited
- Inventors: Yu PENG
- Web page: https://patentgrep.com/patent/US20230397898A1

## CPC

- A61B 6/482: {involving multiple energy imaging}
- A61B 6/032: {Transmission computed tomography [CT]}
- A61B 6/4241: {using energy resolving detectors, e.g. photon counting}
- A61B 6/5217: {extracting a diagnostic or physiological parameter from medical diagnostic data}
- A61B 6/481: {involving the use of contrast agents}
- A61B 6/505: {for diagnosis of bone}
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)

## Abstract

A method and system for generating material decomposition images from plural-energy x-ray based imaging, the method comprising: modelling spatial relationships and spectral relationships among the plurality of images by learning features from the plurality of images in combination and one or more of the plurality of images individually with a deep learning neural network; generating one or more basis material images employing the spatial relationships and the spectral relationships; and generating one or more material specific or material decomposition images from the basis material images. The neural network has an encoder-decoder structure and includes a plurality of encoder branches; each of one or more of the plurality of encoder branches encodes two or more images of the plurality of images in combination; and each of one or more of the plurality of encoder branches encodes a respective individual image of the plurality of images.

## Claims

### Claim 1

(canceled)

### Claim 2

A method of diagnosing, identifying or monitoring a pathology or disease in a subject or patient, the method comprising:
modelling spatial relationships and spectral relationships among a plurality of images of the subject or patient obtained with plural-energy x-ray based imaging, the plurality of images corresponding to respective energies of the plural-energy x-ray based imaging, the modelling comprising learning features from the plurality of images in combination and one or more of the plurality of images individually with a deep learning neural network that has an encoder-decoder structure;
generating one or more basis material images employing the spatial relationships and the spectral relationships;
generating one or more material specific or material decomposition images from the basis material images; and
diagnosing, identifying or monitoring a pathology or disease in the subject or patient from at least one of the generated material specific or material decomposition images.

### Claim 3 (depends on 2)

A method as claimed in claim 2, comprising:
generating any one or more of (i) a bone marrow decomposition image, (ii) a knee cartilage decomposition image, (iii) an iodine contrast decomposition image, (iv) a tumor decomposition image, (v) a muscle and fat decomposition image, (vi) a metal artefact reduction image, and (vii) a beam hardening reduction image.

### Claim 4 (depends on 2)

A method as claimed in claim 2, comprising:
generating one or more bone marrow images, and diagnosing, identifying or monitoring bone marrow related disease using the one or more bone marrow images;
generating one or more knee cartilage images, and diagnosing, identifying or monitoring osteoarthritis or rheumatoid arthritis using the one or more bone marrow images;
generating one or more iodine contrast image, and diagnosing, identifying or monitoring a tumor; and/or
generating one or more muscle images, and diagnosing, identifying or monitoring sarcopenia.

### Claim 5 (depends on 2)

A method as claimed in claim 2, wherein the neural network includes a plurality of encoder branches; each of one or more of the plurality of encoder branches encodes two or more images of the plurality of images in combination; and each of one or more of the plurality of encoder branches encodes a respective individual image of the plurality of images.

### Claim 6 (depends on 5)

The method as claimed in claim 5, wherein:
a) the encoder branches that encode a respective individual image receive in total all of the images that are received in total by the encoder branches that encode two or more images; or
b) the encoder branches that encode a respective individual image encode in total fewer images than are encoded in total by the encoder branches that encode two or more images; or
c) the encoder branches that encode a respective individual image encode in total more images than are encoded in total by the encoder branches that encode two or more images.

### Claim 7

A system for diagnosing, identifying or monitoring a pathology or disease in a subject or patient, the system comprising:
a neural network that has an encoder-decoder structure;
wherein the neural network is configured to model spatial relationships and spectral relationships among a plurality of images of the subject or patient obtained with plural-energy x-ray based imaging, the plurality of images corresponding to respective energies of the plural-energy x-ray based imaging, the modelling comprising learning features from the plurality of images in combination and one or more of the plurality of images individually with a deep learning neural network, and to generate one or more basis material images employing the spatial relationships and the spectral relationships; and
the system is configured to generate one or more material specific or material decomposition images from the basis material images, and to diagnose, identify or monitor a pathology or disease in the subject or patient from at least one of the generated material specific or material decomposition images.

### Claim 8 (depends on 7)

The system as claimed in claim 7, configured to generate any one or more of (i) a bone marrow decomposition image, (ii) a knee cartilage decomposition image, (iii) an iodine contrast decomposition image, (iv) a tumor decomposition image, (v) a muscle and fat decomposition image, (vi) a metal artefact reduction image, and (vii) a beam hardening reduction image.

### Claim 9 (depends on 7)

The system as claimed in claim 7, configured
to generate one or more bone marrow images, and to diagnose, identify or monitor bone marrow related disease using the one or more bone marrow images;
to generate one or more knee cartilage images, and to diagnose, identify or monitor osteoarthritis or rheumatoid arthritis using the one or more bone marrow images;
to generate one or more iodine contrast image, and to diagnose, identify or monitor a tumor; and/or
to generate one or more muscle images, and to diagnose, identify or monitor sarcopenia.

### Claim 10 (depends on 7)

A system as claimed in claim 7, wherein the neural network includes a plurality of encoder branches; each of one or more of the plurality of encoder branches is configured to encode two or more images of the plurality of images in combination; and each of one or more of the plurality of encoder branches is configured to encode a respective individual image of the plurality of images.

### Claim 11 (depends on 10)

The system as claimed in claim 10, wherein:
a) the encoder branches configured to encode a respective individual image encode in total all of the images that are encoded in total by the encoder branches configured to encode two or more images; or
b) the encoder branches that encode a respective individual image are configured to encode in total fewer images than are encoded in total by the encoder branches that encode two or more images; or
c) the encoder branches that encode a respective individual image are configured to encode in total more images than are encoded in total by the encoder branches that encode two or more images.

### Claim 12

A method of improving image quality of an image of a subject or patient, the method comprising:
modelling spatial relationships and spectral relationships among a plurality of images of the subject or patient obtained with plural-energy x-ray based imaging, the plurality of images corresponding to respective energies of the plural-energy x-ray based imaging, the modelling comprising learning features from the plurality of images in combination and one or more of the plurality of images individually with a deep learning neural network that has an encoder-decoder structure;
generating one or more basis material images employing the spatial relationships and the spectral relationships; and
generating one or more material specific or material decomposition images from the basis material images;
generating one or more metal artefact images or beam hardening reduction images; and
improving image quality of at least one of the generated material specific or material decomposition images using the one or more metal artefact or beam hardening reduction images.

### Claim 13 (depends on 12)

The method as claimed in claim 12, comprising:
generating any one or more (a) bone marrow images, (b) knee cartilage images, (c) iodine contrast images, and (d) muscle images; and
improving image quality of the bone marrow, knee cartilage, iodine contrast and/or muscle images using the metal artefact and/or beam hardening reduction images.

### Claim 14 (depends on 12)

The method as claimed in claim 12, wherein the neural network includes a plurality of encoder branches;
each of one or more of the plurality of encoder branches encodes two or more images of the plurality of images in combination;
each of one or more of the plurality of encoder branches encodes a respective individual image of the plurality of images.

### Claim 15 (depends on 14)

The method as claimed in claim 14, wherein:
a) the encoder branches that encode a respective individual image receive in total all of the images that are received in total by the encoder branches that encode two or more images; or
b) the encoder branches that encode a respective individual image encode in total fewer images than are encoded in total by the encoder branches that encode two or more images; or
c) the encoder branches that encode a respective individual image encode in total more images than are encoded in total by the encoder branches that encode two or more images.

### Claim 16

A system improving image quality of an image of a subject or patient, the system comprising:
a neural network that has an encoder-decoder structure;
wherein the neural network is configured to model spatial relationships and spectral relationships among a plurality of images of the subject or patient obtained with plural-energy x-ray based imaging, the plurality of images corresponding to respective energies of the plural-energy x-ray based imaging, the modelling comprising learning features from the plurality of images in combination and one or more of the plurality of images individually with a deep learning neural network, and to generate one or more basis material images employing the spatial relationships and the spectral relationships; and
the system is configured to generate one or more metal artefact images or beam hardening reduction images, and to improve image quality of at least one of the generated material specific or material decomposition images using the one or more metal artefact or beam hardening reduction images.

### Claim 17 (depends on 16)

The system as claimed in claim 16, configured to generate any one or more of (i) a bone marrow decomposition image, (ii) a knee cartilage decomposition image, (iii) an iodine contrast decomposition image, (iv) a tumor decomposition image, (v) a muscle and fat decomposition image, (vi) a metal artefact reduction image, and (vii) a beam hardening reduction image.

### Claim 18 (depends on 16)

The system as claimed in claim 16, wherein the neural network includes a plurality of encoder branches;
each of one or more of the plurality of encoder branches is configured to encode two or more images of the plurality of images in combination; and
each of one or more of the plurality of encoder branches is configured to encode a respective individual image of the plurality of images.

### Claim 19 (depends on 18)

The system as claimed in claim 18, wherein:
a) the encoder branches configured to encode a respective individual image encode in total all of the images that are encoded in total by the encoder branches configured to encode two or more images; or
b) the encoder branches that encode a respective individual image are configured to encode in total fewer images than are encoded in total by the encoder branches that encode two or more images; or
c) the encoder branches that encode a respective individual image are configured to encode in total more images than are encoded in total by the encoder branches that encode two or more images.

### Claim 20

A non-transient computer-readable medium, comprising a computer program comprising program code configured, when executed by one of more computing devices, to implement a method of diagnosing, identifying or monitoring a pathology or disease in a subject or patient, the method comprising:
modelling spatial relationships and spectral relationships among a plurality of images of the subject or patient obtained with plural-energy x-ray based imaging, the plurality of images corresponding to respective energies of the plural-energy x-ray based imaging, the modelling comprising learning features from the plurality of images in combination and one or more of the plurality of images individually with a deep learning neural network that has an encoder-decoder structure;
generating one or more basis material images employing the spatial relationships and the spectral relationships;
generating one or more material specific or material decomposition images from the basis material images; and
diagnosing, identifying or monitoring a pathology or disease in the subject or patient from at least the generated material specific or material decomposition images.

### Claim 21

A non-transient computer-readable medium, comprising a computer program comprising program code configured, when executed by one of more computing devices, to implement a method of improving image quality of an image of a subject or patient, the method comprising:
modelling spatial relationships and spectral relationships among a plurality of images of the subject or patient obtained with plural-energy x-ray based imaging, the plurality of images corresponding to respective energies of the plural-energy x-ray based imaging, the modelling comprising learning features from the plurality of images in combination and one or more of the plurality of images individually with a deep learning neural network that has an encoder-decoder structure;
generating one or more basis material images employing the spatial relationships and the spectral relationships; and
generating one or more material specific or material decomposition images from the basis material images;
generating one or more metal artefact images or beam hardening reduction images; and
improving image quality of at least one of the generated material specific or material decomposition images using the one or more metal artefact or beam hardening reduction images.
