# US 2023/0274476 A1 · Computer-implemented method for determining nuclear medical image data sets in dynamic nuclear medical imaging, determining device and electronically readable storage medium

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

- Number: US 2023/0274476 A1
- Document ID: US20230274476A1
- Status: Published application
- Kind: A1
- Application number: 18/173,951
- Filed: 2023-02-24
- Published: 2023-08-31
- Earliest claimed: 2022-02-28
- Assignees: Siemens Healthcare GmbH
- Applicants: Siemens Healthcare GmbH
- Inventors: Matthias FENCHEL; Thomas Vahle
- Web page: https://patentgrep.com/patent/US20230274476A1

## CPC

- G06T 12/20: Inverse problem, i.e. transformations from projection space into object space
- G06T 12/30: Image post-processing, e.g. metal artefact correction
- A61B 5/055: involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
- G06T 7/0012: {Biomedical image inspection}
- G06T 2207/10088: Magnetic resonance imaging [MRI]
- G06T 2207/10104: Positron emission tomography [PET]
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T 2211/412: Dynamic
- G06T 2211/441: AI-based methods, deep learning or artificial neural networks
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)

## Abstract

Computer-implemented method for determining nuclear medical image data sets in dynamic nuclear medical imaging. The method includes using a trained function to determine at least one further nuclear medical image data set for at least one frame if a basis raw data set is taken from one single frame, wherein input data of the trained function includes at least one of the nuclear medical raw data set of the respective frame or a preliminary reconstructed image reconstructed therefrom, and an already determined nuclear medical image data set.

## Claims

### Claim 1

A computer-implemented method for determining nuclear medical image data sets in dynamic nuclear medical imaging, the method comprising:
providing nuclear medical raw data sets of a field of view acquired over a time interval, wherein the time interval comprises a progression period of a tracer in the field of view and each nuclear medical raw data set is acquired in one of multiple time frames of the time interval;
providing an anatomical image showing an anatomy in the field of view;
selecting at least one of the nuclear medical raw data sets as a basis raw data set;
reconstructing a first nuclear medical image data set by
iterative reconstruction of the basis raw data set based on the anatomical image to incorporate anatomical information, or
using a first trained function applying the iterative reconstruction to the anatomical image and at least one of the basis raw data set or a preliminary reconstructed data set of the basis raw data set; and
using a second trained function to determine at least one further nuclear medical image data set for,
at least one other frame if the basis raw data set is taken from one single frame, wherein input data of the second trained function includes,
at least one of the nuclear medical raw data set of the respective frame or a preliminary reconstructed image reconstructed therefrom, and
an already determined nuclear medical image data set.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
selecting the basis raw data set as a single nuclear medical raw data set already at least one of (i) co-registered to the anatomical image or (ii) from one single frame during which the anatomical image was acquired or closest to the acquisition of the anatomical image.

### Claim 3 (depends on 1)

The method of claim 1, wherein the basis raw data set comprises the nuclear medical raw data sets of all frames.

### Claim 4 (depends on 1)

The method of claim 1, wherein the iterative reconstruction at least one of (i) is based on a maximum likelihood expectation maximization or (ii) uses a Bowsher prior.

### Claim 5 (depends on 1)

The method of claim 1, wherein the first trained function at least one of (i) comprises a convolutional neural network or (ii) uses a preliminary reconstructed image and the anatomical image as input data.

### Claim 6 (depends on 1)

The method of claim 1, wherein the second trained function comprises at least one of a convolutional neural network or a residual network (ResNet).

### Claim 7 (depends on 1)

The method of claim 1, further comprising:
applying attenuation correction based on an attenuation map, the attenuation correction being used in the reconstruction of the first nuclear medical image data set.

### Claim 8 (depends on 7)

The method of claim 7, wherein the applying the attenuation correction includes,
determining the attenuation map from the anatomical image.

### Claim 9 (depends on 7)

The method of claim 7, wherein the input data for the second trained function further includes an attenuation-corrected preliminary reconstructed image based on the attenuation map.

### Claim 10 (depends on 7)

The method of claim 7, wherein
in a case of patient motion, a corrected attenuation map is determined for each frame, to which the second trained function is to be applied, by determining correction parameters from not attenuation corrected preliminary reconstructed images of relevant frames and correcting the attenuation map according to the correction parameters, and
the input data for the second trained function includes an attenuation-corrected preliminary reconstructed image based on the respective corrected attenuation map.

### Claim 11 (depends on 7)

The method of claim 7, wherein the attenuation map is only used for determining the first nuclear medical image data set and the input data for the second trained function includes a not attenuation-corrected preliminary reconstructed image.

### Claim 12 (depends on 1)

The method of claim 1, wherein at least one of the nuclear medical raw data sets are PET raw data sets or the anatomical image is a magnetic resonance image.

### Claim 13

A determining device for determining nuclear medical image data sets in dynamic nuclear medical imaging, comprising:
a first interface configured to provide nuclear medical raw data sets of a field of view acquired over a time interval, wherein the time interval comprises a progression period of a tracer in the field of view and each nuclear medical raw data set is acquired in one of multiple time frames of the time interval;
a second interface configured to provide an anatomical image showing an anatomy in the field of view;
a selection unit configured to select at least one of the nuclear medical raw data sets as a basis raw data set;
a reconstruction unit configured to reconstruct a first nuclear medical image data set by
iterative reconstruction of the basis raw data set based on the anatomical image to incorporate anatomical information, or
using a first trained function applying the iterative reconstruction to the anatomical image and at least one of the basis raw data set or a preliminary reconstructed data set of the basis raw data set;
a determination unit configured to use a second trained function to determine at least one further nuclear medical image data set for,
at least one other frame if the basis raw data set is taken from one single frame, wherein input data of the second trained function includes,
at least one of the nuclear medical raw data set of the respective frame or a preliminary reconstructed image reconstructed therefrom, and
an already determined nuclear medical image data set; and
a third interface configured to output at least the at least one further nuclear medical image data set.

### Claim 14 (depends on 1)

A non-transitory electronically readable storage medium including instructions that, when executed by a determining device of a system, cause the system to perform the method of claim 1.

### Claim 15 (depends on 2)

A non-transitory electronically readable storage medium including instructions that, when executed by a determining device of a system, cause the system to perform the method of claim 2.

### Claim 16 (depends on 5)

The method of claim 5, further comprising:
applying attenuation correction based on an attenuation map, the attenuation correction being used in the reconstruction of the first nuclear medical image data set.

### Claim 17 (depends on 16)

The method of claim 16, wherein the applying the attenuation correction includes,
determining the attenuation map from the anatomical image.

### Claim 18 (depends on 16)

The method of claim 16, wherein the input data for the second trained function further includes an attenuation-corrected preliminary reconstructed image based on the attenuation map.

### Claim 19 (depends on 16)

The method of claim 16, wherein
in a case of patient motion, a corrected attenuation map is determined for each frame, to which the second trained function is to be applied, by determining correction parameters from not attenuation corrected preliminary reconstructed images of relevant frames and correcting the attenuation map according to the correction parameters, and
the input data for the second trained function includes an attenuation-corrected preliminary reconstructed image based on the respective corrected attenuation map.

### Claim 20 (depends on 16)

The method of claim 16, wherein the attenuation map is only used for determining the first nuclear medical image data set and the input data for the second trained function includes a not attenuation-corrected preliminary reconstructed image.
