# US 12,112,447 B2 · Method for identifying, locating and tracking calibrating conduit within 3D ultrasound point cloud

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

- Number: US 12,112,447 B2
- Document ID: US12112447B2
- Status: Granted
- Kind: B2
- Application number: 18/266,831
- Filed: 2023-02-28
- Published: 2024-08-08
- Granted: 2024-10-08
- Earliest claimed: 2022-07-20
- Assignees: Jiangsu Tingsn Technology Co. , Ltd
- Applicants: Jiangsu Tingsn Technology Co., Ltd
- Inventors: Ruifeng Pei; Weijie Qin
- Web page: https://patentgrep.com/patent/US12112447B2

## CPC

- G06T 19/20: Editing of three-dimensional [3D] images, e.g. changing shapes or colours, aligning objects or positioning parts
- A61B 8/0841: {for locating instruments}
- G06T 7/00: Image analysis
- G06T 7/246: using feature-based methods, e.g. the tracking of corners or segments
- G06T 7/277: involving stochastic approaches, e.g. using Kalman filters
- G06T 7/73: using feature-based methods
- G06T 17/00: Three-dimensional [3D] modelling for computer graphics
- G06V 10/34: Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
- G06V 10/44: Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V 10/764: using classification, e.g. of video objects
- G06T 2200/24: involving graphical user interfaces [GUIs]
- G06T 2210/41: Medical
- G06T 2210/56: Particle system, point based geometry or rendering
- G06T 2219/2012: Colour editing, changing, or manipulating; Use of colour codes
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A method for identifying, locating and tracking a calibrating conduit within a 3D ultrasound point cloud includes the following steps: firstly rapidly modeling the conduit through an ultrasound map, then framing an approximate range of the conduit after human-computer interaction in a display interface, and tracking the approximate range accurately. The method has advantages of modeling and locating the conduit based on pure ultrasound, faster and more accurate modeling, more accurate positioning of the conduit by framing, reducing inaccurate positioning caused by other factors, and human-machine interaction for positioning the conduit with ready access to ensure accuracy and reliability. Further, this conduit localization algorithm reduces overall process time and provides a clearer and more real-time image of the conduit in a target environment, eliminating need for an overly demanding target environment and fixed reference points introduced by the conduit, while being more accurate than relying solely on ultrasound modeling.

## Claims

### Claim 1

A method for identifying, localizing and tracking a calibrating conduit within a three-dimensional (3D) ultrasound point cloud, comprising the following steps:
Step S01: modeling an interior of a target environment;
Step S02: performing human-computer interaction to locate a conduit in a user operation interface, and placing the conduit into a scannable range of an ultrasound probe;
Step S03: after starting tracking and placing the conduit into the scannable range of the probe, performing, by the user, human-computer interaction operation on a display interface to frame an area where the conduit is located, and giving content in the frame a color to distinguish therefrom; making, by the user, a determination whether a conduit position is accurate, and if the framing is not correct, the user frames again until the conduit position is accurately identified;
Step S04: after the conduit position is accurately identified, obtaining a range for positioning the conduit, and identifying and tracking the conduit;
Step S05: identifying and obtaining a point cloud of the target conduit, and taking the obtained point cloud as an input to obtain a field point cloud generated from a two-dimensional (2D) ultrasound image by a method for tracking a particle filter tracker; and
Step S06: transmitting new conduit modeling information with the method for tracking the conduit, refreshing a display on a user interface, accurately coloring the conduit with a standard color, and changing a framed color as a standard conduit color and a background color accordingly;
wherein in Step S04, a specific method of obtaining the range for positioning the conduit is as follows: framed content is placed within a 3D frame, and the 3D frame is exactly sized to enclose only the framed content, to obtain a corresponding coordinate range within the 3D frame, which is a range for positioning the conduit.

### Claim 2 (depends on 1)

The method according to claim 1, wherein in Step S03, giving the content selected in the frame a red color, and in Step S06, the standard color is a blue color.

### Claim 3 (depends on 1)

The method according to claim 1, wherein in Step S01, the 2D ultrasound image is scanned periodically with help of continuous rotation in space, and the interior of the target environment is modeled using 3D modeling means.

### Claim 4 (depends on 3)

The method according to claim 3, wherein in Step S05, a specific method for identifying the point cloud of the conduit is as follows:
Step S51, in an initialization step, downsampling a framed point cloud to reduce an amount of data;
Step S52: extracting a Scale Invariant Feature Transform (SIFT) key point of the point cloud;
Step S53: then calculating Signature of Histogram of Orientation (SHOT) features for the SIFT key points, establishing a local coordinate system at the feature points, and combining spatial location information of neighboring points with geometric feature statistical information to describe the feature points; and
Step S54: extracting the features, generating a feature vector, and inputting the feature vector into a support vector machine (SVM) classifier to obtain the point cloud of the conduit.

### Claim 5 (depends on 3)

The method according to claim 3, wherein in Step S03, giving the content selected in the frame a red color, and in Step S06, the standard color is a blue color.

### Claim 6 (depends on 1)

The method according to claim 1, wherein in Step S03, a framing of impurities outside the conduit is reduced provided that framing is performed.

### Claim 7 (depends on 6)

The method according to claim 6, wherein in Step S05, a specific method for identifying the point cloud of the conduit is as follows:
Step S51, in an initialization step, downsampling a framed point cloud to reduce an amount of data;
Step S52: extracting a Scale Invariant Feature Transform (SIFT) key point of the point cloud;
Step S53: then calculating Signature of Histogram of Orientation (SHOT) features for the SIFT key points, establishing a local coordinate system at the feature points, and combining spatial location information of neighboring points with geometric feature statistical information to describe the feature points; and
Step S54: extracting the features, generating a feature vector, and inputting the feature vector into a support vector machine (SVM) classifier to obtain the point cloud of the conduit.

### Claim 8 (depends on 6)

The method according to claim 6, wherein in Step S03, giving the content selected in the frame a red color, and in Step S06, the standard color is a blue color.

### Claim 9

A method for identifying, localizing and tracking a calibrating conduit within a three-dimensional (3D) ultrasound point cloud, comprising the following steps:
Step S01: modeling an interior of a target environment;
Step S02: performing human-computer interaction to locate a conduit in a user operation interface, and placing the conduit into a scannable range of an ultrasound probe;
Step S03: after starting tracking and placing the conduit into the scannable range of the probe, performing, by the user, human-computer interaction operation on a display interface to frame an area where the conduit is located, and giving content in the frame a color to distinguish therefrom; making, by the user, a determination whether a conduit position is accurate, and if the framing is not correct, the user frames again until the conduit position is accurately identified;
Step S04: after the conduit position is accurately identified, obtaining a range for positioning the conduit, and identifying and tracking the conduit;
Step S05: identifying and obtaining a point cloud of the conduit, and taking the obtained point cloud as an input to obtain a field point cloud generated from a two-dimensional (2D) ultrasound image by a method for tracking a particle filter tracker; and
Step S06: transmitting new conduit modeling information with the method for tracking the conduit, refreshing a display on a user interface, accurately coloring the conduit with a standard color, and changing a framed color as a standard conduit color and a background color accordingly;
wherein in Step S05, a specific method for identifying the point cloud of the conduit is as follows:
Step S51, in an initialization step, downsampling a framed point cloud to reduce an amount of data;
Step S52: extracting a Scale Invariant Feature Transform (SIFT) key point of the point cloud;
Step S53: then calculating Signature of Histogram of Orientation (SHOT) features for the SIFT key points, establishing a local coordinate system at the feature points, and combining spatial location information of neighboring points with geometric feature statistical information to describe the feature points; and
Step S54: extracting the features, generating a feature vector, and inputting the feature vector into a support vector machine (SVM) classifier to obtain the point cloud of the conduit.

### Claim 10 (depends on 9)

The method according to claim 9, wherein in Step S05, a point cloud obtained by an identifying module needs to be pre-processed by:
Step S61: utilizing a direct-through filter, retaining the point cloud in an effective region, specifying ranges in x, y, and z latitudes, respectively, and filtering out the point cloud outside the range; and
Step S62: downsampling the filtered point cloud to reduce the amount of data.

### Claim 11 (depends on 10)

The method according to claim 10, wherein in Step S03, giving the content selected in the frame a red color, and in Step S06, the standard color is a blue color.

### Claim 12 (depends on 10)

The method according to claim 10, wherein in Step S62, a sampling method is geometric sampling.

### Claim 13 (depends on 12)

The method according to claim 12, wherein in Step S03, giving the content selected in the frame a red color, and in Step S06, the standard color is a blue color.