Device and method for detecting guidewire

Patent metadata

Field	Value
Publication number	US-2022378514-A1
Application number	US-202217752344-A
Country	US
Kind code	A1
Filing date	May 24, 2022
Priority date	May 25, 2021
Publication date	Dec 1, 2022
Grant date	—

Abstract

An electronic device includes a processor configured to generate a position movement prediction field indicating prediction of a potential positional change of a branch path by a patient's biological activity for one or more branch paths based on a blood vessel image of a reference frame, correct guidewire information extracted from a blood vessel image of a target frame with respect to a catheter position of the reference frame, and select a branch path to dispose the guidewire information, among one or more branch paths of a blood vessel region based on the position movement prediction field and the corrected guidewire information; and a display configured to visualize the guidewire information on the selected branch path.

First claim

What is claimed is:

1 . An electronic device, comprising:
a processor configured to generate a position movement prediction field indicating prediction of a potential positional change of a branch path by a patient's biological activity for one or more branch paths based on a blood vessel image of a reference frame, correct guidewire information extracted from a blood vessel image of a target frame with respect to a catheter position of the reference frame, and select a branch path to dispose the guidewire information, among one or more branch paths of a blood vessel region based on the position movement prediction field and the corrected guidewire information; and a display configured to visualize the guidewire information on the selected branch path.

2 . The electronic device of claim 1 , further comprising:
an image acquiring unit configured to acquire coronary angiographic images of the reference frame and the target frame as blood vessel images of the reference frame and the target frame.

3 . The electronic device of claim 1 , wherein the processor generates an entire path length map indicating a path length to points in a segmented blood vessel region from the blood vessel image of the reference frame and generates the position movement prediction field for each of one or more branch paths of the blood vessel region based on the entire path length map.

4 . The electronic device of claim 3 , wherein the processor extracts a blood vessel region image in which the blood vessel region is segmented from the blood vessel image of the reference frame, detects a catheter position from the blood vessel image of the reference frame based on a first machine learning model to set the catheter position as a reference position, and generates an entire path length map indicating a path length from a start position set using the reference position to points in the blood vessel region.

5 . The electronic device of claim 3 , wherein the processor acquires a branch path length map for each of one or more branch paths from the generated entire path length map and generates a position movement prediction field of the corresponding branch path based on the branch path length map for each of the one or more branch paths.

6 . The electronic device of claim 1 , wherein the processor determines the same value as the path length value of the corresponding point in the position movement prediction field for each branch path, as a predictive value of a position predicted that each point in the corresponding branch path is potentially moved by a biological activity.

7 . The electronic device of claim 6 , wherein the processor determines each predictive value of the position movement prediction field based on Euclidean distance transform between each point in the branch path and a point outside the branch path.

8 . The electronic device of claim 1 , wherein the processor detects a catheter position of the reference frame from the blood vessel image of the reference frame based on a first machine learning model, detects a catheter position of the target frame from the blood vessel image of the target frame based on the first machine learning model, and corrects the guidewire information based on a difference between the catheter position of the reference frame and the catheter position of the target frame.

9 . The electronic device of claim 8 , wherein the processor extracts the guidewire information from the blood vessel image of the target frame based on a second machine learning model which is different from the first machine learning model.

10 . The electronic device of claim 1 , wherein the processor extracts a guidewire candidate region for the one or more branch paths based on the corrected guidewire information and the generated position movement prediction field and selects one branch path among the one or more branch paths, based on comparison between the guidewire candidate region for the one or more branch paths and the guidewire information.

11 . The electronic device of claim 10 , wherein the processor acquires a plurality of predictive values corresponding to a guidewire in the corresponding branch path by mapping the corrected guidewire information on the position movement prediction field for each of the one or more branch paths and extracts the guidewire candidate region of the corresponding branch path based on the plurality of acquired predictive values.

12 . The electronic device of claim 11 , wherein the processor extracts a guidewire candidate region of the corresponding branch path based on a minimum value and a maximum value among the plurality of acquired predictive values.

13 . The electronic device of claim 10 , wherein the processor calculates a distance score between the guidewire candidate region for the one or more branch paths and the guidewire information and selects a branch path having a candidate region indicating a highest distance score, among the guidewire candidate regions.

14 . The electronic device of claim 10 , wherein the processor selects a branch path based on a comparison history between the guidewire candidate region of each branch path in a previous frame of the target frame and the guidewire information and a comparison result between the guidewire candidate region of each branch path in the target frame and the guidewire information.

15 . The electronic device of claim 14 , wherein the processor applies a first weight to a first comparison score in a first previous frame which precedes the target frame by one frame difference, applies a second weight which is different from the first weight to a second comparison score in a second previous frame which precedes the target frame by a second frame difference which is larger than the first frame difference, and selects the branch path using a fused result value based on a first partial score obtained by applying the first weight to the first comparison score and a second partial score obtained by applying the second weight to the second comparison score.

16 . The electronic device of claim 1 , wherein the processor fits the guidewire information to a center line of the selected branch path and the display overlays the fitted guidewire information onto the branch path.

17 . The electronic device of claim 1 , wherein the target frame is a frame after the reference frame.

18 . The electronic device of claim 1 , wherein the reference frame is a frame corresponding to a timing when a contrast is injected into a patient.

19 . A guidewire detecting method performed by an electronic device, the method comprising:
generating a position movement prediction field indicating prediction of a potential positional change of a branch path by a patient's biological activity for one or more branch paths based on a blood vessel image of a reference frame; correcting guidewire information extracted from a blood vessel image of a target frame with respect to a catheter position of the reference frame; selecting a branch path to dispose the guidewire information, among one or more branch paths of a blood vessel region based on the position movement prediction field and the corrected guidewire information; and visualizing the guidewire information on the selected branch path.

20 . A computer program which is stored in a computer readable medium to be coupled to hardware to execute the method of claim 19 .

Assignees

• Asan Found
• Univ Ulsan Found Ind Coop
• Medipixel Inc

Inventors

• Kweon Jihoon
• Kim Young Eon
• Song Kyo Seok

Classifications

• A61B2017/00703

  correcting for movement of heart, e.g. ECG-triggered · CPC title

• G06T7/248

  involving reference images or patches · CPC title

• A61B6/5294

  involving using additional data, e.g. patient information, image labeling, acquisition parameters · CPC title

• A61B6/5217

  extracting a diagnostic or physiological parameter from medical diagnostic data · CPC title

• A61B6/481

  involving the use of contrast agents · CPC title