Granted Patent B2

US 12,667,436 · App. 18/492,743 · Granted Jun 30, 2026

Inventors: Dong Wu (Wuhan, CN); Shaowen He (Wuhan, CN); Yunhong Yang (Wuhan, CN); Guoqiang Wang (Wuhan, CN)

Assignee: WUHAN UNITED IMAGING SURGICAL CO., LTD.

A61B34/30A61B34/10A61B34/25B25J9/1664G05B19/4155G06T7/11G06T7/174G06T7/337A61B2034/101A61B2034/105A61B2034/107G05B2219/50391G06N20/00G06T2207/20021G06T2207/20092G06T2207/30004

Abstract

Claims (95)

1 . A surgical pathway processing system, comprising an image segmentation module, an avoidance region determination module, and a pathway planning module, wherein the surgical pathway processing system performs operations including:

obtaining, by the image segmentation module, a first segmented image by performing image segmentation on a first medical image;

determining, by the avoidance region determination module, a region to be avoided based on the first segmented image; and

determining, by the pathway planning module, a surgical pathway based on the region to be avoided,

wherein

the obtaining the first segmented image includes: obtaining an overall mask, a predetermined object mask, and a to-be-intervened object mask by segmenting the first medical image, wherein the to-be-intervened object mask refers to a mask of a region to be intervened during an interventional procedure, and

the determining the region to be avoided includes determining the region to be avoided based on the overall mask, the predetermined object mask, and the to-be-intervened object mask.

2 . The surgical pathway processing system of claim 1 , wherein the determining the region to be avoided based on the overall mask, the predetermined object mask, and the to-be-intervened object mask includes:

determining a target mask based on the overall mask and the predetermined object mask; and

determining the region to be avoided based on the target mask and the to-be-intervened object mask.

3 . The surgical pathway processing system of claim 2 , wherein

the determining the target mask based on the overall mask and the predetermined object mask includes:

determining the target mask by subtracting the predetermined object mask from the overall mask;

determining a non-intervening object mask by subtracting the to-be-intervened object mask from the target mask; and

determining the region to be avoided includes:

performing a connected component analysis on the non-intervening object mask; and

determining a connected component satisfying a first predetermined condition as the region to be avoided.

4 . The surgical pathway processing system of claim 1 , wherein the obtaining the first segmented image by performing the image segmentation on the first medical image includes:

determining a target medical image by performing a registration using finite elements based on the first segmented image.

5 . The surgical pathway processing system of claim 4 , wherein the determining the target medical image by performing the registration using the finite elements based on the first segmented image includes:

obtaining a second medical image of a target object, the second medical image and the first medical image including images of the target object at different periods;

obtaining a second segmented image by segmenting at least a portion of the second medical image;

obtaining a rigid registration result by performing a rigid registration on the target object in the first segmented image and the target object in the second segmented image;

obtaining an elastic registration result by performing an elastic registration on the target object in the rigid registration result and the target object in the second segmented image; and

obtaining the target medical image by processing the first segmented image based on the rigid registration result and the elastic registration result.

6 . The surgical pathway processing system of claim 5 , the operations further including:

constructing a mechanical model of the target object based on the target object in the rigid registration result;

obtaining a surface elasticity registration result by performing a surface elasticity registration on the target object in the first segmented image and the target object in the second segmented image;

obtaining a deformation field of each node of the mechanical model of the target object by solving the mechanical model using a difference between the rigid registration result of the target object and the surface elasticity registration result of the target object as a boundary condition; and

obtaining the target medical image by processing the first segmented image based on a rigid transformation matrix in the rigid registration result and the deformation field of each node of the mechanical model of the target object.

7 . The surgical pathway processing system of claim 6 , the operations further including:

obtaining a transformed first medical image by performing translation and/or rotation on the first medical image based on the rigid transformation matrix; and

obtaining the target medical image by performing translation and/or rotation on each node of the mechanical model of the target object in the transformed first medical image based on the deformation field of each node of the mechanical model of the target object.

8 . The surgical pathway processing system of claim 5 , the operations further including:

obtaining a third medical image of the target object, the third medical image including an enhanced type of image, the second medical image including a plain scanned type image obtained after image transformation of the third medical image.

9 . The surgical pathway processing system of claim 4 , the operations further including:

obtaining a third medical image of a target object, wherein the third medical image includes an enhanced type image, the third medical image and the first medical image including images of the target object at different periods;

obtaining a third mask image corresponding to the third medical image by extracting the target object and an internal detail mask from the third medical image;

obtaining a fourth mask image corresponding to the target medical image by extracting the target object and an internal detail mask from the target medical image;

calculating a count of duplicate pixel points in a region of interest in the third mask image and a region of interest in the fourth mask image;

obtaining a validation value based on the count of duplicate pixel points, a count of pixel points in the region of interest in the third mask image, and a count of pixel points in the region of interest in the fourth mask image; and

if the verification value is greater than a predetermined threshold, determining that a verification of a registered image passes, and if the verification value is not greater than the predetermined threshold, determining that the verification of the registered image does not pass.

10 . The surgical pathway processing system of claim 1 , wherein the determining the surgical pathway based on the region to be avoided includes:

automatically determining a target point and an intervention point based on a user input; and

determining the surgical pathway based on the target point, the intervention point, and the region to be avoided.

11 . The surgical pathway processing system of claim 10 , wherein the automatically determining the target point and the intervention point based on the user input includes:

receiving a first operation of a user on the first segmented image, and determining the target point in response to the first operation;

receiving a second operation of the user on the first segmented image, and determining a reference point in response to the second operation;

determining a reference pathway based on the target point and the reference point, the reference pathway including a straight line connecting the target point and the reference point; and

determining the intervention point based on the reference pathway.

12 . The surgical pathway processing system of claim 10 , wherein the automatically determining the intervention point based on the user input includes:

determining an action pathway of a second operation of a user on the first segmented image;

determining a reference pathway based on the action pathway; and

determining the intervention point based on the reference pathway.

13 . The surgical pathway processing system of claim 11 , the operations further including:

determining a candidate pathway based on the reference pathway; and

determining the surgical pathway by verifying the candidate pathway.

14 . The surgical pathway processing system of claim 13 , the operations further including:

determining whether the candidate pathway includes an interference feature and/or whether an intervention parameter of the candidate pathway satisfies a second predetermined condition; and

if the candidate pathway does not include the interference feature and/or the intervention parameter of the candidate pathway satisfies the second predetermined condition, determining that a verification result of the candidate pathway is passed; or if the candidate pathway includes the interference feature and/or the intervention parameter of the candidate pathway does not satisfy the second predetermined condition, determining that the verification result of the candidate pathway is not passed.

15 . The surgical pathway processing system of claim 14 , the operations further including:

selecting a point in the candidate pathway that is a predetermined distance away from the target point as a verification point;

obtaining a voxel pixel value of the verification point in a predetermined neighborhood, and determining whether the voxel pixel value includes a value in a predetermined pixel set;

if the voxel pixel value includes a value in the predetermined pixel set, determining that the candidate pathway includes the interference feature; and

if the voxel pixel value does not include a value in the predetermined pixel set, updating the verification point in the candidate pathway until a voxel pixel value of each verification point in the candidate pathway in a corresponding predetermined neighborhood does not include a value in the predetermined pixel set, and determining that the candidate pathway does not include the interference feature.

16 . The surgical pathway processing system of claim 14 , the operations further including:

determining a pathway length and/or an intervention angle of the candidate pathway;

determining whether the pathway length is less than or equal to a predetermined length threshold and/or determining whether the intervention angle is less than or equal to a predetermined angle threshold;

if the pathway length is less than or equal to the predetermined length threshold, and/or the intervention angle is less than or equal to the predetermined angle threshold, determining that the intervention parameter of the candidate pathway satisfies the second predetermined condition; and if the pathway length is greater than the predetermined length threshold, and/or the intervention angle is greater than the predetermined angle threshold, determining that the intervention parameter of the candidate pathway does not satisfy the second predetermined condition; wherein

the second predetermined condition includes the predetermined length threshold and the predetermined angle threshold.

17 . The surgical pathway processing system of claim 1 , wherein the determining the surgical pathway based on the region to be avoided includes:

obtaining a distance threshold for the region to be avoided based on a surgical procedure; and

determining the surgical pathway based on a target point, an intervention point, the region to be avoided, and the distance threshold.

18 . A surgical pathway processing method performed by the surgical pathway processing system of claim 1 , comprising:

obtaining the first segmented image by performing the image segmentation on the first medical image;

determining the region to be avoided based on the first segmented image; and

determining the surgical pathway based on the region to be avoided,

wherein

the obtaining the first segmented image includes obtaining an overall mask, a predetermined object mask, and a to-be-intervened object mask by segmenting the first medical image, wherein the to-be-intervened object mask refers to a mask of a region to be intervened during an interventional procedure; and

the determining the region to be avoided includes determining the region to be avoided based on the overall mask, the predetermined object mask, and the to-be-intervened object mask.

19 . A surgical pathway processing device, comprising:

a display device;

a surgical robot; and

a control device, comprising one or more processors and a storage, wherein the storage includes operation instructions configured to direct the one or more processors to perform operations including:

obtaining a first segmented image by performing image segmentation on a first medical image displayed by the display device;

determining a region to be avoided based on the first segmented image;

determining a surgical pathway based on the region to be avoided; and

driving the surgical robot to move based on the surgical pathway,

wherein

the obtaining the first segmented image includes obtaining an overall mask, a predetermined object mask, and a to-be-intervened object mask by segmenting the first medical image, wherein the to-be-intervened object mask refers to a mask of a region to be intervened during an interventional procedure; and

the determining the region to be avoided includes determining the region to be avoided based on the overall mask, the predetermined object mask, and the to-be-intervened object mask.

20 . The surgical pathway processing system of claim 1 , the operations further including:

obtaining a to-be-avoided object mask by segmenting the first medical image;

performing a connected component analysis on the to-be-avoided object mask; and

determining a connected component satisfying a first predetermined condition as the region to be avoided.

Assignments (2)

Priority Claims (4)

Continuity (2)

References Cited (104)