# US 2021/0241520 A1 · Systems, methods, and computer-readable media for automatic computed tomography to computed tomography registration

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

- Number: US 2021/0241520 A1
- Document ID: US20210241520A1
- Status: Published application
- Kind: A1
- Application number: 17/233,817
- Filed: 2021-04-19
- Published: 2021-08-05
- Earliest claimed: 2017-12-13
- Assignees: Covidien LP
- Applicants: Covidien LP
- Inventors: Jiguang SHEN; Lev A. Koyrakh; Oren P. Weingarten; Ron Barak
- Web page: https://patentgrep.com/patent/US20210241520A1

## CPC

- G06T 17/00: Three-dimensional [3D] modelling for computer graphics
- G06T 5/20: using local operators
- G06T 7/0012: {Biomedical image inspection}
- G06T 7/344: {involving models}
- G06T 7/62: of area, perimeter, diameter or volume
- G06T 7/64: of convexity or concavity
- G06T 2207/10081: Computed x-ray tomography [CT]
- G06T 2207/30061: Lung
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

Systems, methods, and computer-readable media for registering initial computed tomography (CT) images of a luminal network with subsequent CT images of the luminal network include obtaining initial CT images of the luminal network and subsequent CT images of the luminal network, generating an initial three-dimensional (3D) model of the luminal network based on the initial CT images of the luminal network, generating a subsequent 3D model of the luminal network based on the subsequent CT images of the luminal network, and matching the initial 3D model with the subsequent 3D model based on a registration.

## Claims

### Claim at position 1

1-20. (canceled)

### Claim 21

A system for registering initial images of a luminal network with subsequent images of the luminal network, the system comprising:
a computing device including a processor and a memory; and
an application, stored in the memory and executed by the processor, which when executed, causes the computing device to:
generate an initial model of a luminal network based on initial images of the luminal network;
generate a subsequent model of the luminal network based on subsequent images of the luminal network;
match the initial model with the subsequent model;
for each of the initial and subsequent models of the luminal network:
deconstruct the luminal network into a plurality of generations, each generation of the plurality of generations originating at a generation bifurcation of the luminal network and having a subset of airway branches extending from the generation bifurcation;
generate, for each airway branch of a respective subset of airway branches of each generation, a curved line segment; and
determine a curvature of each curved line segment; and
match a plurality of curved line segments of each subset of airway branches of the initial model with a plurality of curved line segments of each subset of airway branches of the subsequent model.

### Claim 22 (depends on 21)

The system according to claim 21, wherein the application, when executed, causes the computing device to apply, for each curved line segment, a linear interpolant connecting a first reference point along a respective curved line segment to a second reference point along the respective curved line segment.

### Claim 23 (depends on 22)

The system according to claim 22, wherein the application, when executed, causes the computing device to determine the curvature of each curved line segment based on a respective linear interpolant.

### Claim 24 (depends on 21)

The system according to claim 21, wherein the application, when executed, causes the computing device to compare, for each subset of airway branches of each generation, at least one curvature of each curved line segment of the initial model to at least one curvature of each curved line segment of the subsequent model.

### Claim 25 (depends on 24)

The system according to claim 24, wherein matching of the plurality of curved line segments of each subset of airway branches of the initial model with the plurality of curved line segments of each subset of airway branches of the subsequent model is based on the comparison.

### Claim 26 (depends on 21)

The system according to claim 21, wherein generating, for each airway branch of a respective subset of airway branches of each generation, a curved line segment is based on a connected sequence of line segments originating at a respective generation bifurcation and extending along a respective airway branch.

### Claim 27 (depends on 21)

The system according to claim 21, wherein in matching the plurality of curved line segments of each subset of airway branches of the initial model with the plurality of curved line segments of each subset of airway branches of the subsequent model, the application when executed, further causes the computing device to:
apply at least one of piecewise derivative overlapping filter, a curve similarity filter, a length similarity filter, or an angle similarity filter; and
unmatch, based on the piecewise derivative overlapping filter, the curve similarity filter, the length similarity filter, or the angle similarity filters, portions of the initial model and of the subsequent model that were matched in the matching of the initial model with the subsequent model.

### Claim 28 (depends on 21)

The system according to claim 21, wherein in matching the plurality of curved line segments of each subset of airway branches of the initial model with the plurality of curved line segments of each subset of airway branches of the subsequent model, the application when executed, further causes the computing device to:
apply at least one of a local displacement consistency filter, a patch transformation consistency filter, or an isomorphism scoring system; and
unmatch, based on the local displacement consistency filter, the patch transformation consistency filter, or the isomorphism scoring system, portions of the initial model and of the subsequent model that were matched in the matching of the initial model with the subsequent model.

### Claim 29 (depends on 21)

The system according to claim 21, wherein the initial and subsequent images are CT images.

### Claim 30 (depends on 21)

The system according to claim 21, wherein the application, when executed, causes the computing device to receive the initial images of the luminal network and the subsequent images of the luminal network.

### Claim 31 (depends on 21)

The system according to claim 21, wherein in matching the initial model with the subsequent model, the application when executed, further causes the computing device to:
select a plurality of initial reference points within the initial model; and
select a plurality of subsequent reference points within the subsequent model,
wherein the plurality of initial reference points and the plurality of subsequent reference points include x, y, and z coordinates.

### Claim 32 (depends on 31)

The system according to claim 31, wherein the matching of the initial model with the subsequent model includes determining an alignment between the initial model and the subsequent model that minimizes a divergence between the plurality of initial reference points and the plurality of subsequent reference points.

### Claim 33 (depends on 31)

The system according to claim 31, wherein the application, when executed, further cause the computing device to:
select fiducial points within the plurality of initial reference points and fiducial points within the plurality of subsequent reference points;
match the fiducial points within the plurality of subsequent reference points with the fiducial points within the plurality of initial reference points to generate a plurality of point pairs;
generate a plurality of additional point pairs; and
derive a thin plate splines transformation based on the plurality of point pairs and the plurality of additional point pairs.

### Claim 34

A system for registering initial images of a luminal network with subsequent of the luminal network, the system comprising:
a computing device including a processor and a memory; and
an application, stored in the memory and executed by the processor, which when executed, causes the computing device to:
generate an initial model of a luminal network based on initial images of the luminal network;
generate a subsequent model of the luminal network based on subsequent images of the luminal network;
match the initial model with the subsequent model;
for each of the initial and subsequent models of the luminal network:
deconstruct the luminal network into a plurality of generations, each generation of the plurality of generations originating at a generation bifurcation of the luminal network and having a subset of airway branches extending from the generation bifurcation;
generate, for each airway branch of a respective subset of airway branches of each generation, a curved line segment; and
determine a curvature of each curved line segment;
compare, for each subset of airway branches of each generation, at least one curvature of each curved line segment of the initial model to at least one curvature of each curved line segment of the subsequent model; and
unmatch portions of the initial model from portions of the subsequent model based on the comparison.

### Claim 35 (depends on 34)

The system according to claim 34, wherein generating the curved line segment for each airway branch of a respective subset of airway branches of each generation is based on a connected sequence of line segments originating at a respective generation bifurcation and extending along a respective airway branch.

### Claim 36 (depends on 34)

The system according to claim 34, wherein the application, when executed, causes the computing device to apply, for each curved line segment, a linear interpolant connecting a first reference point along a respective curved line segment to a second reference point along the respective curved line segment.

### Claim 37 (depends on 36)

The system according to claim 36, wherein determining of the curvature of each curved line segment is based on a respective linear interpolant.

### Claim 38 (depends on 34)

The system according to claim 34, wherein the application, when executed, causes the computing device to register the initial model with the subsequent model based on portions of the initial and subsequent models that remain matched after the unmatching.

### Claim 39 (depends on 34)

The system according to claim 34, wherein the application, when executed, causes the computing device to receive the initial images of the luminal network and the subsequent images of the luminal network.

### Claim 40

A system for registering initial images of a luminal network with subsequent images of the luminal network, the system comprising:
a computing device including a processor and a memory; and
an application, stored in the memory and executed by the processor, which when executed, causes the computing device to:
receive initial images of the luminal network;
receive subsequent images of the luminal network;
for each of the initial and subsequent images of the luminal network:
deconstruct the luminal network into a plurality of generations, each generation of the plurality of generations originating at a generation bifurcation of the luminal network and having a subset of airway branches extending from the generation bifurcation;
generate, for each airway branch of a respective subset of airway branches of each generation, a curved line segment; and
determine a curvature of each curved line segment; and
match a plurality of curved line segments of each subset of airway branches of the initial images with a plurality of curved line segments of each subset of airway branches of the subsequent images.