Computer-aided-design of skeletal implants

1. A computer-aided design method for producing an implant for a patient prior to operation, the method comprising:

generating digital image data with a non-invasive three-dimensional (3D) volume scan of an anatomical region of the patient comprising a defect site that is to receive the implant;

generating a polygonal-mesh 3D surface image representing the anatomical region of the patient based on the digital image data;

defining a defect margin crestline describing the external boundary of the patient's defect based on a curvature estimation on the polygonal-mesh 3D surface image;

generating a digital template of desired anatomy associated with the anatomical region;

performing a two-pass thin plate spline (TPS) warp on the digital template to register the digital template specifically to the patient;

generating a prototype implant surface based on the defect margin crestline and the registered template of the anatomical region, the prototype implant surface defining an outer surface of a 3D digital model of the implant;

generating at least one internal surface and at least one connecting surface with respect to the prototype implant surface to generate the 3D digital model of the implant; and

fabricating the implant based on the 3D digital model of the implant.

2. The method of claim 1 , wherein the polygonal-mesh 3D surface image is of an entirety of the anatomical region that is represented by the digital image data, the method further comprising:

determining a 3D region of interest (ROI) on the polygonal-mesh 3D surface image of the anatomical region that includes only a region of the polygonal-mesh 3D surface image immediately adjacent to and surrounding the defect site; and

extracting the 3D ROI from the polygonal-mesh 3D surface image to generate a polygonal-mesh 3D surface image file of the 3D ROI;

wherein defining the defect margin crestline comprises defining the defect margin crestline on the 3D ROI.

3. The method of claim 2 , wherein determining the 3D ROI comprises:

placing a plurality of seed landmarks in the region of the polygonal-mesh 3D surface image immediately adjacent to the defect site such that the seed landmarks are coplanar to define a best-fitting plane;

connecting consecutive ones of the plurality of seed landmarks based on a line equation; and

defining the 3D ROI as a 3D area of the polygonal-mesh 3D surface image that is enclosed within the connected plurality of seed landmarks.

4. The method of claim 1 , wherein defining the defect margin crestline comprises:

determining a 3D surface curvature at each respective 3D surface point of the polygonal-mesh 3D surface image to generate the curvature estimate; and

generating color values corresponding to a magnitude of the 3D surface curvature at each respective 3D surface point of the polygonal-mesh 3D surface image.

5. The method of claim 4 , wherein determining the 3D surface comprises:

defining an arc length between predefined starting and ending points defining a specified arc length for a parametric space curve along the polygonal-mesh 3D surface image; and

calculating a unit vector perpendicular to a tangent of the parametric space curve at each respective 3D surface point.

6. The method of claim 4 , further comprising redistributing the color values in a histogram over an entire intensity scale to ascertain contrasts between low and high curvature of the respective 3D surface points.

7. The method of claim 4 , further comprising:

placing seed landmarks at 3D surface points having highest curvature magnitudes based on the color values;

forming sectors on the polygonal-mesh 3D surface image based on the seed landmarks; and

executing a minimum-cost path traversal algorithm in each sector to determine a series of points defining the defect margin crestline.

8. The method of claim 1 , wherein generating the digital template of the anatomical region comprises generating the digital template based on one of left-right mirroring of the anatomical region and providing a digital template that is an average of the anatomical region imaged on a plurality of patients.

9. The method of claim 8 , wherein generating the digital template comprises implementing left-right mirroring of the anatomical region in response to the defect site being unilateral with respect to a plane separating left and right sides of the patient.

10. The method of claim 8 , wherein generating the digital template comprises providing an average of the anatomical region that is an appropriate size and reoriented for a reference frame of the patient in response to the defect site spanning a plane separating left and right sides of the patient.

11. The method of claim 1 , wherein generating the at least one internal surface and the at least one connecting surface comprises extending the prototype implant surface such that it spans a thickness of a body structure of the patient underlying the defect margin crestline based on the polygonal-mesh 3D surface image.

12. The method of claim 11 , wherein extending the prototype implant surface comprises:

casting geometric rays from each point of the defect margin crestline internally in a normal direction of a best-fitting plane of seed landmarks that defines a 3D region of interest (ROI) of the anatomical region that includes the defect site; and

calculating a Euclidean distance between an origin of each geometric ray and the best-fitting plane.

13. The method of claim 1 , wherein the prototype implant surface is configured as a triangular-mesh isosurface image, the method further comprising:

resampling the prototype implant surface to uniformly shape and space triangles associated with the triangular-mesh isosurface image;

implementing a surface smoothing algorithm to remove slice-edge artifact from the resampled prototype implant surface; and

implementing an articulation algorithm to articulate the surface-smoothed and resampled prototype implant surface to fit the defect margin crestline.

14. The method of claim 13 , wherein implementing the surface smoothing algorithm comprises removing slice edge artifact in response to volume image slices having an inter-slice dimension that is larger than an in-slice pixel dimension.

15. The method of claim 1 , wherein generating the at least one internal surface and the at least one connecting surface comprises generating connecting surfaces between the prototype implant surface and the internal surface with a taper-fit surface that is tapered between the prototype implant surface and the internal surface.

16. The method of claim 15 , further comprising validating the 3D digital model of the implant with respect to contact of the taper-fit surface and the internal surface with adjacent bone and soft tissue structures prior to fabricating the implant.

17. The method of claim 16 , wherein validating the 3D digital model of the implant comprises implementing a collision detection algorithm to verify seating of the taper-fit surface with adjacent bone structure and to verify non-collision with adjacent had and soft tissues.

18. The method of claim 1 , further comprising selecting the defect site of the patient from at least one of a hard tissue element and a soft tissue element.

19. The method of claim 1 , further comprising selecting the defect site of the patient as a cranial defect associated with the skull of the patient.