Deformable registration of computer-generated airway models to airway trees
A system for registering a luminal network to a 3D model of the luminal network includes a computing device configured to identify potential matches in the 3D model with location data of a location sensor, assigning one of the potential matches a registration score based on a deformation model applied to the 3D model, and displaying the potential match having the highest registration score.
1. A method of registering a luminal network to a 3D model of the luminal network, the method comprising:
generating a three-dimensional (3D) model from a plurality of images of a luminal network;
receiving location data associated with a location sensor on a tool navigating the luminal network;
identifying a plurality of potential matches in the 3D model with the location data;
applying a deformation model to at least a portion of the 3D model;
assigning registration scores to the potential matches in the 3D model with the location data based on the deformation model; and
displaying one of the plurality of potential matches having a highest registration score.
2. The method of claim 1 , wherein the assigning of the registration score is further based on at least one of:
a distance between the potential match and the location data; or
a difference in direction between location sensor direction data and a direction of the potential match.
3. The method of claim 1 , further comprising:
receiving further location data; and
reassigning the one of the plurality of potential matches an updated registration score based on the further location data.
4. The method of claim 1 , wherein anchor points in the luminal network are used to constrain the registration.
5. The method of claim 1 , further comprising discarding any of the plurality of potential matches assigned a low registration score.
6. The method of claim 5 , further comprising delaying discarding potential matches with low registration scores when the potential matches are in a bifurcation area of the 3D model.
7. The method of claim 1 , further comprising:
identifying a first region of the luminal network and a second region of the luminal network, wherein the deformation model includes at least one of:
modeling at least one of rotation, compression, extension, or bending for the first region and the second region independently;
modeling at least one of rotation, compression, extension, or bending for the first region and the second region with adjacent regions having interdependence; or
performing rigid registration of the first and second regions to form first and second rigid registrations, and stitching together the first and second rigid registrations.
8. The method of claim 1 , wherein:
the luminal network is a bronchial airway; and
the deformation model comprises applying a moving window from a trachea to a periphery of the bronchial airway, and performing a rigid registration in the moving window.
9. The method of claim 1 , wherein the deformation model comprises at least one of:
weighing registration for central areas of the luminal network greater than peripheral areas of the luminal network;
using deformation models having different deformation characteristics for at least two of a right lung, a left lung, or each lung lobe of the luminal network;
using a biomechanical model of the luminal network based on finite image analysis;
using a deformation model based on a specific disease; or
using at least one of a bronchoscope model and a catheter model to produce a specific deformation model.
10. A system for registering a luminal network to a 3D model of the luminal network, the system comprising:
a location sensor on a tool configured to be navigated within a luminal network;
a tracking system configured to detect a location of the location sensor within luminal network; and
a computing device including a processor and a memory storing instructions which, when executed by the processor, cause the computing device to:
generate a three-dimensional (3D) model from a plurality of images of a luminal network;
receive location data associated with the location sensor on the tool navigating the luminal network;
identify a plurality of potential matches in the 3D model with the location data;
apply a deformation model to at least a portion of the 3D model;
assign registration scores to the potential matches in the 3D model with the location data based on the deformation model; and
display one of the plurality of potential matches having a highest registration score.
11. The system of claim 10 , wherein the assigning of the registration score is further based on at least one of:
a distance between the potential match and the location data; or
a difference in direction between location sensor direction data and a direction of the potential match.
12. The system of claim 10 , further comprising instructions which when executed by the processor, cause the computing device to:
receive further location data and reassign at least one of the potential matches an updated registration score based on the further location data.
13. The system of claim 10 , wherein the instructions when executed by the processor, cause the computing device to:
identify a first region of the luminal network and a second region of the luminal network, wherein the deformation model executes steps of at least one of:
model at least one of rotation, compression, extension, or bending for the first region and the second region independently;
model at least one of rotation, compression, extension, or bending for the first region and the second region with adjacent regions having interdependence; or
perform rigid registration of the first and second regions to form first and second rigid registrations, and stitching together the first and second rigid registrations.
14. The system of claim 10 , wherein the luminal network is a patient's lung airways, wherein the deformation model weighs registration for central areas of the lungs greater than peripheral areas of the lungs.
15. The system of claim 10 , wherein the instructions when executed by the processor, cause the computing device to:
employ a deformation model having different deformation characteristics for one or more of a right lung, a left lung, or each lung lobe.
16. The system of claim 10 , wherein the deformation model is one or more of:
a biomechanical model of a patient's lungs based on finite image analysis;
a deformation model based on a specific disease; or
a catheter navigation based model.
17. A system for registering a luminal network to a 3D model of the luminal network, the system comprising:
a location sensor on a tool configured to be navigated within a luminal network;
a tracking system configured to detect a location of the location sensor within luminal network; and
a computing device including a processor and a memory storing instructions which, when executed by the processor, cause the computing device to:
generate a three-dimensional (3D) model from a plurality of images of a luminal network;
receive location data associated with the location sensor on the tool navigating the luminal network;
identify a plurality of potential matches in the 3D model with the location data;
apply a deformation model to at least a portion of the 3D model;
assign registration scores to the potential matches in the 3D model with the location data based on the deformation model, wherein the assigning of the registration score is further based on at least one of:
a distance between the potential match and the location data; or
a difference in direction between location sensor direction data and a direction of the potential match; and
display one of the plurality of potential matches having a highest registration score.
18. The system of claim 17 , wherein the instructions when executed by the processor, cause the computing device to:
identify a first region of the luminal network and a second region of the luminal network, wherein the deformation model executes steps of at least one of:
model at least one of rotation, compression, extension, or bending for the first region and the second region independently;
model at least one of rotation, compression, extension, or bending for the first region and the second region with adjacent regions having interdependence; or
perform rigid registration of the first and second regions to form first and second rigid registrations, and stitching together the first and second rigid registrations.
19. The system of claim 17 , wherein the luminal network is the airways of a patient's lungs, wherein the deformation model weighs registration for central areas of the lungs greater than peripheral areas of the lungs.
20. The system of claim 17 , wherein the instructions when executed by the processor, cause the computing device to:
employ a deformation model having different deformation characteristics for one or more of a right lung, a left lung, or each lung lobe.