Ultrasound based multiple bone registration surgical systems and methods of use in computer-assisted surgery

Patent metadata

Field	Value
Publication number	US-12544137-B2
Application number	US-202117452317-A
Country	US
Kind code	B2
Filing date	Oct 26, 2021
Priority date	Oct 27, 2020
Publication date	Feb 10, 2026
Grant date	Feb 10, 2026

Abstract

Disclosed herein is a surgical system for surgical registration of patient bones to a surgical plan. As part of its ability to perform the surgical registration, the system is configured to process an ultrasound image of patient bones, the ultrasound image including a bone surface for each of the patient bones. The system includes a computing device including a processing device and a computer-readable medium with one or more executable instructions stored thereon. The processing device is configured to execute the one or more executable instructions. The one or more executable instructions: i) detects the bone surface of each of the patient bones in the ultrasound image; and ii) segregates a first point cloud of ultrasound image pixels associated with the bone surface of each of the patient bones.

First claim

What is claimed is:

1 . A surgical system configured to process an ultrasound image of intraoperative patient bones, the ultrasound image including a bone surface for each of the intraoperative patient bones, the system comprising:
at least one surgical tool; and a computing device including a processing device and a non-transitory computer-readable medium with one or more executable instructions stored thereon, the processing device configured to execute the one or more executable instructions, the one or more executable instructions including: i) detecting the bone surface of each of the intraoperative patient bones in the ultrasound image as ultrasound image pixels; ii) segregating a first point cloud of the ultrasound image pixels associated with the bone surface of each of the intraoperative patient bones; iii) compute a transformation of the first point cloud into a segregated 3D point cloud that is segregated such that the ultrasound image pixels of the segregated 3D point cloud are each correlated to a corresponding bone surface of the intraoperative patient bones; iv) computing an initial or rough registration of a second point cloud taken from the intraoperative patient bones to pre-operative patient specific computer models of patient bones; and v) compute a final multiple bone registration employing the initial or rough registration and the segregated 3D point cloud, wherein the final multiple bone registration achieves a final registration between the segregated 3D point cloud and the intraoperative patient bones, wherein the at least one surgical tool is in communication with the computing device and the final multiple bone registration serves as an input to navigate the at least one surgical tool in relation to each of the intraoperative patient bones.

2 . The system of claim 1 , wherein the detecting of the bone surfaces occurs via an image processing algorithm forming at least a portion of the one or more executable instructions.

3 . The system of claim 2 , wherein the image processing algorithm includes a machine learning model.

4 . The system of claim 2 , wherein the segregating of the first point cloud occurs via a pixel classification neural network forming at least a portion of the one or more executable instructions.

5 . The system of claim 2 , wherein the segregating of the first point cloud occurs via an image-based classification neural network forming at least a portion of the one or more executable instructions.

6 . The system of claim 1 , wherein, in computing the transformation of the first point cloud into the segregated 3D point cloud, the ultrasound image pixels are calibrated to an ultrasound probe tracker and the ultrasound probe tracker is calibrated to a tracking camera.

7 . The system of claim 6 , wherein calibrating the ultrasound image pixels to the ultrasound probe tracker is based on a propagation speed of ultrasound waves in a certain medium.

8 . The system of claim 1 , wherein, in computing the transformation of the first point cloud into the segregated 3D point cloud, the ultrasound image pixels are calibrated to an ultrasound probe tracker, wherein the ultrasound probe tracker is calibrated to one or more of a tracking camera and a coordinate system relative to the bone surface via an anatomy tracker located on the bone surface of the intraoperative patient bones.

9 . The system of claim 1 , wherein the segregating of the first point cloud occurs via geometric analysis of the first point cloud.

10 . The system of claim 1 , wherein the second point cloud includes multiple point clouds relative to multiple trackers on the intraoperative patient bones.

11 . The system of claim 10 , wherein the multiple point clouds include one point cloud registered to one computer model of the pre-operative patient specific computer models of the patient bones and another point cloud registered to another computer model of the pre-operative patient specific computer models of the patient bones.

12 . The system of claim 1 , wherein the initial or rough registration is landmark based.

13 . The system of claim 1 , wherein the initial or rough registration is computed from a position and orientation of anatomy trackers.

14 . The system of claim 1 , wherein, in computing the initial or rough registration, a third point cloud and a fourth point cloud are generated by the system, the third point cloud being of a first bone of the intraoperative patient bones relative to a first tracker associated with the first bone, the fourth point cloud being of a second bone of the intraoperative patient bones relative to a second tracker associated with the second bone.

15 . The system of claim 14 , wherein, in computing the initial or rough registration, the system matches bony surface points of the third point cloud onto a computer model of the first bone and the system matches bony surface points of the fourth point cloud onto a computer model of the second bone.

16 . The system of claim 1 , wherein, in computing the final multiple bone registration wherein there is the final registration between the segregated 3D point cloud and the intraoperative patient bones, the system iteratively refines registration of the segregated 3D point cloud to the pre-operative patient specific computer models of the patient bones, and iteratively refines the segregation of the segregated 3D point cloud.

17 . A method of processing an ultrasound image of intraoperative patient bones, the ultrasound image including a bone surface for each of the intraoperative patient bones, the method comprising:
detecting the bone surface of each of the intraoperative patient bones in the ultrasound image as ultrasound image pixels; segregating a first point cloud of the ultrasound image pixels associated with the bone surface of each of the intraoperative patient bones; computing a transformation of the first point cloud into a segregated 3D point cloud that is segregated such that the ultrasound image pixels of the segregated 3D point cloud are each correlated to a corresponding bone surface of the intraoperative patient bones; computing an initial or rough registration of a second point cloud taken from the patient bones to pre-operative patient specific computer models of patient bones; computing a final multiple bone registration employing the initial or rough registration and the segregated 3D point cloud, wherein the final multiple bone registration achieves a final registration between the segregated 3D point cloud and the intraoperative patient bones; and using the final multiple bone registration to navigate at least one surgical tool in relation to each of the intraoperative patient bones.

18 . The method of claim 17 , wherein the detecting of the bone surfaces occurs via an image processing algorithm.

19 . The method of claim 18 , wherein the image processing algorithm includes a machine learning model.

20 . The method of claim 18 , wherein the segregating of the first point cloud occurs via a pixel classification neural network.

21 . The method of claim 18 , wherein the segregating of the first point cloud occurs via an image-based classification neural network.

22 . The method of claim 17 , wherein, in computing the transformation of the first point cloud into the segregated 3D point cloud, the ultrasound image pixels are calibrated to an ultrasound probe tracker and the ultrasound probe tracker is calibrated to a tracking camera.

Assignees

• Mako Surgical Corp

Inventors

• Zimmermann Peter
• Prevost Raphael

Classifications

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title

• A61B2034/2063

  Acoustic tracking systems, e.g. using ultrasound · CPC title

• A61B2034/2055

  Optical tracking systems · CPC title

• G06T2207/10028

  Range image; Depth image; 3D point clouds · CPC title

• G06T2207/20081

  Training; Learning · CPC title