Computer-aided tracking and motion analysis with ultrasound for measuring joint kinematics

1. A motion analysis system, comprising:

at least one ultrasonic imaging device;

at least one motion sensing device;

at least one computing device in data communication with the at least one ultrasonic imaging device and the at least one motion sensing device; and

a tracking application executable in the at least one computing device, the tracking application comprising logic that:

accesses video generated from the at least one ultrasonic imaging device and the at least one motion sensing camera;

tracks a target patch embodied in a plurality of frames of the ultrasound video by:

compressing a plurality of patches of individual ones of the frames of the ultrasound video into a plurality of vectors;

generating a space partitioning data structure for each of the frames of the ultrasound video;

identifying an image intensity feature for each frame utilizing a corresponding one of the space partitioning data structures generated for each frame of the ultrasound video; and

determines a plurality of optimized tracking locations for a sequence of the ultrasound video using the image intensity feature identified for each frame.

2. The system of claim 1 , wherein the space partitioning data structure further comprises a binary space partitioning data tree.

3. The system of claim 2 , wherein the binary space partitioning data tree further comprises a k-dimensional tree.

4. The system of claim 3 , wherein compressing the patches of the individual ones of the frames of the ultrasound video into the vectors further comprises applying principal components analysis (PCA) to obtain a plurality of principal components from a set of randomly selected ones of the patches.

5. The system of claim 4 , wherein the k-dimensional tree is generated using the principal components.

6. The system of claim 3 , wherein the tracking application further comprises logic that clusters a plurality of similarly tracked nodes in the k-dimensional tree.

7. The system of claim 1 , wherein determining the plurality of optimized tracking locations further comprises applying dynamic programming.

8. The system of claim 1 , wherein the tracking application is executed to track joint kinematics for a joint of a person or an animal.

9. The system of claim 8 , wherein the target patch comprises a target point located on an image of a greater trochanter of the person or the animal.

10. The system of claim 1 , further comprising a display in data communication with the at least one computing device.

11. The system of claim 10 , wherein the at least one ultrasonic imaging device, the at least one motion sensing device, the display, and the at least one computing device are implemented in a single portable device.

12. A non-transitory computer-readable medium embodying a program executable in at least one computing device, comprising code that:

accesses an ultrasound video generated by an ultrasound device in communication with the at least one computing device;

tracks a target patch embodied in a plurality of frames of the ultrasound video by:

compressing a plurality of patches of individual ones of the frames of the ultrasound video into a plurality of vectors;

generating a space partitioning data structure for each of the frames of the ultrasound video;

identifying an image intensity feature for each frame utilizing a corresponding one of the space partitioning data structures generated for each frame of the ultrasound video; and

determining a plurality of optimized tracking locations.

13. The non-transitory computer-readable medium of claim 12 , wherein the space partitioning data structure further comprises a binary space partitioning data tree.

14. The non-transitory computer-readable medium of claim 13 , wherein the binary space partitioning data tree further comprises a k-dimensional tree.

15. The non-transitory computer-readable medium of claim 14 , wherein compressing the patches of the individual ones of the frames of the ultrasound video into the vectors further comprises applying principal components analysis (PCA) to obtain a plurality of principal components from a set of randomly selected ones of the patches.

16. The non-transitory computer-readable medium of claim 15 , wherein the k-dimensional tree is generated using the principal components.

17. The non-transitory computer-readable medium of claim 14 , wherein the program further comprises code that clusters a plurality of similarly tracked nodes in the k-dimensional tree.

18. The non-transitory computer-readable medium of claim 12 , wherein determining the plurality of optimized tracking locations further comprises applying dynamic programming.

19. A computer-implemented method, comprising:

accessing, by at least one computing device, an ultrasound video generated from at least one ultrasonic imaging device and at least one motion sensing camera;

tracking, by the at least one computing device, a target patch embodied in a plurality of frames of the ultrasound video by:

compressing a plurality of patches of individual ones of the frames of the ultrasound video into a plurality of vectors;

generating a space partitioning data structure for each of the frames of the ultrasound video;

identifying an image intensity feature for each frame utilizing a corresponding one of the space partitioning data structures generated for each frame of the ultrasound video; and

determines a plurality of optimized tracking locations for a sequence of the ultrasound video using the image intensity feature identified for each frame.

20. The computer-implemented method of claim 19 , wherein the space partitioning data structure further comprises a binary space partitioning data tree.

21. The computer-implemented method of claim 20 , wherein the binary space partitioning data tree further comprises a k-dimensional tree.

22. The computer-implemented method of claim 21 , wherein compressing the patches of the individual ones of the frames of the ultrasound video into the vectors further comprises applying principal components analysis (PCA) to obtain a plurality of principal components from a set of randomly selected ones of the patches.

23. The computer-implemented method of claim 22 , wherein the k-dimensional tree is generated using the principal components.

24. The computer-implemented method of claim 21 , further comprising clustering, by the at least one computing device, a plurality of similarly tracked nodes in the k-dimensional tree.

25. The computer-implemented method of claim 19 , wherein determining, by the at least one computing device, the plurality of optimized tracking locations further comprises applying, by the at least one computing device, dynamic programming.