Systems and methods of registration for image-guided procedures

1. A device, comprising:

an instrument usable to collect location data for one or more passageways; and

one or more processors coupled to the instrument;

wherein the one or more processors are configured to:

organize a plurality of points within the location data by clustering the plurality of points to generate a plurality of clusters based on distances between points of the plurality of points;

create a passageway tree based on the organized points;

identify at least three non-collinear landmark locations within the passageway tree;

create a seed transformation between one or more of the at least three non-collinear landmark locations and corresponding model locations in model data; and

register, using the seed transformation, the plurality of points to the model data for the one or more passageways.

2. The device of claim 1 , wherein the instrument is a flexible catheter comprising a shape sensor, wherein the location data is provided by the shape sensor.

3. The device of claim 2 , wherein the shape sensor is a fiber optic shape sensor.

4. The device of claim 1 , further comprising a tracking sensor coupled to the instrument.

5. The device of claim 1 , wherein the plurality of points are clustered within corresponding bins, and wherein the distances between the points are distances between the points in each of the corresponding bins.

6. The device of claim 1 , wherein to create the passageway tree the one or more processors are further configured to:

iterate through the plurality of points in descending order of insertion depth, wherein respective insertions depths are based on a corresponding insertion depth of the instrument when each of the plurality of points is collected; and

create parent-child relationships between nodes of the passageway tree corresponding to the points based on distances between the points.

7. The device of claim 6 , wherein to create the passageway tree the one or more processors are further configured to link a first node corresponding to a first point as a first parent to a second node corresponding to a second point closest to the first point that also has a greater insertion depth than the first point.

8. The device of claim 6 , wherein to create the passageway tree the one or more processors are further configured to link a first node corresponding to a first point as a parent to each of a second node and a third node, the second and third node corresponding to two closest points to the first point that also have a greater insertion depth than the first point.

9. The device of claim 6 , wherein to create the passageway tree the one or more processors are further configured to create a leaf node corresponding to a first point when the first point is not within a threshold distance of any point having a greater insertion depth than the first point.

10. The device of claim 1 , wherein to identify the at least three non-collinear landmark locations within the passageway tree, the one or more processors are further configured to:

identify a main branch point in the passageway tree;

identify a first landmark location proximal to the main branch point;

identify a second landmark location distal to the main branch point in a first branch of the passageway tree distal to the main branch point; and

identify a third landmark location distal to the main branch point in a second branch of the passageway tree distal to the main branch point, the second branch being different from the first branch.

11. The device of claim 10 , wherein the one or more processors are further configured to determine a location of the main branch point based on an aggregation of the location data corresponding to each of the points within a threshold distance of the main branch point.

12. The device of claim 10 , wherein the one or more processors are further configured to determine a location of the first landmark location based on the location data corresponding to a point located at a first insertion depth proximal to the main branch point, the first insertion depth being a desired distance proximal from a second insertion depth of the main branch point.

13. The device of claim 10 , wherein the one or more processors are further configured to determine a location of the first landmark location based on an aggregation of the location data corresponding to each of the points having a first insertion depth within a range of desired distances proximal to a second insertion depth of the main branch point.

14. The device of claim 10 , wherein the one or more processors are further configured to determine a location of the second landmark location based on the location data corresponding to a point in the first branch located at a first insertion depth distal to the main branch point, the first insertion depth being a desired distance distal from a second insertion depth of the main branch point.

15. The device of claim 10 , wherein the one or more processors are further configured to determine whether the first branch is a left branch or a right branch in the passageway tree based on an order in which the first branch is traversed, steering commands of the instrument recorded when the location data is collected, one or more of an orientation angle and a length of the first branch, or a known left-right orientation of a sensor system used by the instrument to collect the location data.

16. The device of claim 10 , wherein the main branch point is a most proximal node of the passageway tree having two children.

17. A method of registration using one or more processors, the method comprising:

collecting a set of sensor data during insertion of a flexible elongate device within a plurality of passageways, wherein the sensor data comprises a plurality of points representing a plurality of locations of the flexible elongate device within the plurality of passageways;

organizing the plurality of points by clustering the plurality of points to generate a plurality of clusters based on distances between points of the plurality of points;

creating a passageway tree based on the organized points;

identifying at least three non-collinear landmark locations within the passageway tree;

creating a seed transformation between one or more of the at least three non-collinear landmark locations and corresponding model locations in model data; and

registering, using the seed transformation, the plurality of points to the model data for the plurality of passageways.

18. The method of claim 17 , wherein identifying the at least three non-collinear landmark locations comprises:

identifying a main branch point in the passageway tree;

identifying a first landmark location proximal to the main branch point;

identifying a second landmark location distal to the main branch point in a first branch of the passageway tree, wherein the first branch is distal to the main branch point; and

identifying a third landmark location distal to the main branch point in a second branch of the passageway tree, wherein the second branch is distal to the main branch point, the second branch being different from the first branch.

19. The method of claim 17 , wherein creating the passageway tree includes:

iterating through the plurality of points in descending order of insertion depth, wherein respective insertions depths are based on a corresponding insertion depth of the flexible elongate device when each of the plurality of points is collected; and

creating parent-child relationships between nodes of the passageway tree corresponding to the points based on distances between the points.

20. A non-transitory machine-readable medium comprising a plurality of machine-readable instructions which when executed by one or more processors associated with a device are adapted to cause the one or more processors to perform operations including:

collecting a set of sensor data during insertion of a flexible elongate device within a plurality of passageways, wherein the sensor data comprises a plurality of points representing a plurality of locations of the flexible elongate device within the plurality of passageways;

organizing the plurality of points by clustering the plurality of points to generate a plurality of clusters based on distances between points of the plurality of points;

creating a passageway tree based on the organized points;

identifying at least three non-collinear landmark locations within the passageway tree;

creating a seed transformation between one or more of the at least three non-collinear landmark locations and corresponding model locations in model data; and

registering, using the seed transformation, the plurality of points to the model data for the plurality of passageways.