# US 2025/0209643 A1 · Systems, devices, and methods for image registration

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

- Number: US 2025/0209643 A1
- Document ID: US20250209643A1
- Status: Published application
- Kind: A1
- Application number: 19/001,617
- Filed: 2024-12-26
- Published: 2025-06-26
- Earliest claimed: 2023-12-26
- Assignees: Shanghai United Imaging Healthcare Co., Ltd.
- Applicants: Shanghai United Imaging Healthcare Co., Ltd.
- Inventors: Xiaowei LYU; Xin WENG
- Web page: https://patentgrep.com/patent/US20250209643A1

## CPC

- G06T 7/30: Determination of transform parameters for the alignment of images, i.e. image registration
- G06T 7/33: using feature-based methods
- G06T 7/0012: {Biomedical image inspection}
- G06T 7/12: Edge-based segmentation
- G06T 7/246: using feature-based methods, e.g. the tracking of corners or segments
- G06T 7/73: using feature-based methods
- G06T 2200/04: involving 3D image data
- G06T 2207/10016: Video; Image sequence
- G06T 2207/10121: Fluoroscopy
- G06T 2207/30101: Blood vessel; Artery; Vein; Vascular
- G06T 2207/30172: Centreline of tubular or elongated structure
- G06T 2207/30241: Trajectory
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

The present disclosure provides systems, devices, and methods for image registration. The methods include obtaining a first medical image and a second medical image of a target subject, the first medical image indicating blood vessels of the target subject, the second medical image indicating an interventional tool in the target subject; determining an initial registration relationship between the first medical image and the second medical image; determining, based on a first physiological feature of the target subject at an acquisition time of the second medical image, a first motion state of the blood vessels at the acquisition time; and determining a target registration relationship between the first medical image and the second medical image based on the initial registration relationship and the first motion state of the blood vessels.

## Claims

### Claim 1

A method for image registration, implemented on a computing device having at least one storage device storing a set of instructions, and at least one processor in communication with the at least one storage device, the method comprising:
obtaining a first medical image and a second medical image of a target subject, the first medical image indicating blood vessels of the target subject, the second medical image indicating an interventional tool in the target subject;
determining an initial registration relationship between the first medical image and the second medical image;
determining, based on a first physiological feature of the target subject at an acquisition time of the second medical image, a first motion state of the blood vessels at the acquisition time; and
determining a target registration relationship between the first medical image and the second medical image based on the initial registration relationship and the first motion state of the blood vessels.

### Claim 2 (depends on 1)

The method of claim 1, wherein the first medical image includes an angiography image collected before a surgery, and the second medical image includes a fluoroscopy image collected in the surgery.

### Claim 3 (depends on 1)

The method of claim 1, wherein the determining an initial registration relationship between the first medical image and the second medical image includes:
determining centerlines of the blood vessels in the first medical image;
determining a centerline of the interventional tool in the second medical image; and
determining the initial registration relationship based on the centerlines of the blood vessels and the centerline of the interventional tool.

### Claim 4 (depends on 3)

The method of claim 3, wherein the determining the initial registration relationship based on the centerlines of the blood vessels and the centerline of the interventional tool includes:
obtaining projection data of the centerlines of the blood vessels by projecting the centerlines to a plane corresponding to the second medical image;
determining a target centerline that matches the centerline of the interventional tool based on the projection data of the centerlines; and
determining the initial registration relationship by performing a registration operation on the target centerline and the centerline of the interventional tool.

### Claim 5 (depends on 1)

The method of claim 1, wherein the determining a first motion state of the blood vessels at the acquisition time includes:
obtaining a mapping relationship between physiological phases and blood vessel motion states; and
determining the first motion state of the blood vessels based on the first physiological feature and the mapping relationship.

### Claim 6 (depends on 5)

The method of claim 5, wherein the mapping relationship is determined by:
obtaining sample medical images of the target subject and a sample physiological feature corresponding to each sample medical image, the sample medical images being collected sequentially over at least one physiological cycle and indicating the blood vessels of the target subject;
determining relative motions of the blood vessels between pairs of adjacent sample medical images among the sample medical images; and
determining the mapping relationship based on the sample physiological feature and the relative motions of the blood vessels.

### Claim 7 (depends on 6)

The method of claim 6, wherein the determining relative motions of the blood vessels between pairs of adjacent sample medical images among the sample medical images includes:
for each pair of adjacent sample medical images,
determining position information of a region of interest (ROI) in each sample medical image in the pair; and
determining the relative motion of the blood vessels based on the position information of the ROI in the sample medical images in the pair.

### Claim 8 (depends on 6)

The method of claim 6, wherein the determining the mapping relationship based on the sample physiological feature and the relative motions of the blood vessels includes:
for each physiological phase in the at least one physiological cycle,
selecting, from the sample medical images, one or more sample medical images corresponding to the physiological phase based on the sample physiological feature corresponding to each sample medical image;
determining the blood vessel motion state corresponding to the physiological phase based on the relative motions corresponding to the one or more selected sample medical images; and
determining the mapping relationship based on the blood vessel motion state corresponding to each physiological phase.

### Claim 9 (depends on 6)

The method of claim 6, wherein the sample medical images include angiography images or fluoroscopy images.

### Claim 10 (depends on 5)

The method of claim 5, wherein the mapping relationship is determined based on sample medical images of a sample subject different from the target subject, the sample medical images being collected sequentially over at least one physiological cycle and indicating blood vessels of the sample subject.

### Claim 11 (depends on 1)

The method of claim 1, wherein the determining a target registration relationship based on the initial registration relationship and the first motion state of the blood vessels includes:
generating a first auxiliary registration relationship between the first medical image and the second medical image based on the first motion state of the blood vessels;
determining a first weight for the initial registration relationship and a second weight for the first auxiliary registration relationship based on the second medical image;
determining the target registration relationship based on the first auxiliary registration relationship, the initial registration relationship, the first weight, and the second weight.

### Claim 12 (depends on 11)

The method of claim 11, wherein the determining a first weight for the initial registration relationship and a second weight for the first auxiliary registration relationship based on the second medical image includes:
segmenting a centerline of the interventional tool in the second medical image;
determining a first evaluation result of a segmentation result of the centerline of the interventional tool; and
determining the first weight and the second weight based on the first evaluation result.

### Claim 13 (depends on 11)

The method of claim 11, further comprising:
determining a physiological phase corresponding to the second medical image based on the first physiological feature, wherein the first weight and the second weight are further determined based on the physiological phase corresponding to the second medical image.

### Claim 14 (depends on 11)

The method of claim 11, further comprising:
determining a second evaluation result of the noise level in the second medical image, wherein the first weight and the second weight are further determined based on the second evaluation result.

### Claim 15 (depends on 1)

The method of claim 1, wherein the determining a target registration relationship comprises:
obtaining a second physiological feature of the target subject at the acquisition time of the second medial image;
determining a second motion state of the blood vessels based on the second physiological feature; and
determining the target registration relationship based on the initial registration relationship, the first motion state, and the second motion state.

### Claim 16 (depends on 1)

The method of claim 1, wherein the second medical image is collected during a surgery, and the determining the target registration relationship comprises:
obtaining monitoring information of the body of the target subject during the surgery;
in response to detecting a rigid motion of the body of the target subject based on the monitoring information, updating the initial registration relationship based on the detected rigid motion;
determining the target registration relationship based on the updated initial registration relationship and the first motion state.

### Claim 17 (depends on 1)

The method of claim 1, wherein the determining a first motion state comprises:
obtaining a first physiological curve of the target subject in a pre-scan and sample medical images of the target subject collected sequentially over multiple physiological cycles in the pre-scan;
obtaining a second physiological curve of the target subject during a period including the acquisition time of the second medical image;
identifying a target portion in the first physiological curve that matches the second physiological curve; and
determining the first motion state based on the sample medical images corresponding to the target portion in the first physiological curve.

### Claim 18

A system for image registration, comprising:
at least one storage device including a set of instructions; and
at least one processor configured to communicate with the at least one storage device, wherein when executing the set of instructions, the at least one processor is configured to direct the system to perform operations including:
obtaining a first medical image and a second medical image of a target subject, the first medical image indicating blood vessels of the target subject, the second medical image indicating an interventional tool in the target subject;
determining an initial registration relationship between the first medical image and the second medical image;
determining, based on a first physiological feature of the target subject at an acquisition time of the second medical image, a first motion state of the blood vessels at the acquisition time; and
determining a target registration relationship between the first medical image and the second medical image based on the initial registration relationship and the first motion state of the blood vessels.

### Claim 19 (depends on 18)

The system of claim 18, wherein the first medical image includes an angiography image collected before a surgery, and the second medical image includes a fluoroscopy image collected in the surgery.

### Claim 20

A method for visualizing blood vessels in medical images, implemented on a computing device having at least one storage device storing a set of instructions, and at least one processor in communication with the at least one storage device, the method comprising:
obtaining a 3D angiography image of a patient before a surgery, the 3D angiography image indicating blood vessels of the patient;
obtaining multiple 2D fluoroscopy images of the patient in the surgery in real time;
obtaining physiological motion information of the patient synchronously with the multiple 2D fluoroscopy images;
for each 2D fluoroscopy image in the multiple 2D fluoroscopy images, obtaining a real-time motion trajectory of the blood vessels in the surgery based on the 2D fluoroscopy image and the physiological motion information;
generating a 3D blood vessel model based on the 3D angiography image; and
generating a visualized blood vessel image by performing, based on the real-time motion trajectory and the 2D fluoroscopy image, a spatial transformation operation on the 3D blood vessel model or a 2D projection image of the 3D blood vessel model.
