# US 12,217,442 B2 · Systems and methods for registration between patient space and image space using registration frame apertures

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,217,442 B2
- Document ID: US12217442B2
- Status: Granted
- Kind: B2
- Application number: 17/650,053
- Filed: 2022-02-04
- Published: 2022-08-11
- Granted: 2025-02-04
- Earliest claimed: 2021-02-10
- Assignees: Medtronic Navigation, Inc.
- Applicants: Medtronic Navigation, Inc.
- Inventors: Ryan D. Datteri; Ganesh Saiprasad; Steven Hartmann; Matthew F. Dicorleto; Brandon Merkl
- Web page: https://patentgrep.com/patent/US12217442B2

## CPC

- A61B 6/12: Arrangements for detecting or locating foreign bodies
- G06T 7/337: {involving reference images or patches}
- A61B 6/032: {Transmission computed tomography [CT]}
- A61B 34/20: Surgical navigation systems; Devices for tracking or guiding surgical instruments, e.g. for frameless stereotaxis
- G06T 3/14: Transformations for image registration, e.g. adjusting or mapping for alignment of images
- G06T 7/0014: {using an image reference approach}
- G06T 7/174: involving the use of two or more images
- G06T 7/248: {involving reference images or patches}
- G06T 7/68: of symmetry
- G06T 7/73: using feature-based methods
- G06T 7/74: {involving reference images or patches}
- A61B 2017/00725: {Calibration or performance testing}
- A61B 2034/2055: {Optical tracking systems}
- A61B 2034/2065: {Tracking using image or pattern recognition}
- A61B 2090/3762: {using computed tomography systems [CT]}
- A61B 2090/3937: {Visible markers}
- A61B 2090/3983: {Reference marker arrangements for use with image guided surgery}
- A61B 2090/3991: {having specific anchoring means to fixate the marker to the tissue, e.g. hooks}
- A61B 2090/502: {Headgear, e.g. helmet, spectacles}
- G06T 2207/10072: Tomographic images
- G06T 2207/10081: Computed x-ray tomography [CT]
- G06T 2207/20036: Morphological image processing
- G06T 2207/20132: Image cropping
- G06T 2207/20224: Image subtraction
- G06T 2207/30004: Biomedical image processing
- G06T 2207/30012: Spine; Backbone
- G06T 2207/30016: Brain
- G06T 2207/30052: Implant; Prosthesis
- G06T 2207/30204: Marker
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

Systems and methods for performing registration between a patient space and an image space are disclosed herein. Systems using such methods may identify a pose of a registration frame having one or more apertures within the patient space using a tracking system. Image data corresponding to the image space having an image of the registration frame is taken. A plurality of aperture locations within the image data (corresponding to the apertures of the registration frame) are identified in the image data. Aperture representations from a model of the registration frame are matched to these aperture locations to determine a pose of the registration frame within the image space. A transform between the patient space and the image space is generated based on the pose of the registration frame within the patient space and the pose of the registration frame within the image space. Optimization methods for this transform are discussed.

## Claims

### Claim 1

A method of performing registration between a patient space and an image space, comprising:
imaging the patient space to generate image data of the image space, the image data comprising an image of a registration frame that is located in the patient space, the registration frame comprising a plurality of apertures;
identifying a plurality of registration frame aperture locations within the image data, each of the plurality of aperture locations within the image data respectively corresponding to one of the plurality of apertures of the registration frame;
identifying a pose of the registration frame within the patient space using a tracking system for the patient space;
determining a pose of the registration frame within the image space using the plurality of registration frame aperture locations within the image data; and
computing a transform between the patient space and the image space based on the pose of the registration frame within the patient space and the pose of the registration frame within the image space;
wherein the image of the registration frame comprises high intensity voxels corresponding to the material of the registration frame and low intensity voxels corresponding to each of the plurality of apertures of the registration frame, and wherein identifying the plurality of registration frame aperture locations within the image data using the image of the registration frame comprises:
performing a closing operation on the image of the registration frame to generate a closed image of the registration frame having high intensity voxels corresponding to the material of the registration frame and high intensity voxels corresponding to each of the plurality of apertures of the registration frame;
subtracting the image of the registration frame from the closed image of the registration frame to generate an inverted image of the registration frame comprising low intensity voxels corresponding to the material of the registration frame and high intensity voxels corresponding to each of the plurality of apertures of the registration frame; and
identifying centroids of groupings of the high intensity voxels in the inverted image of the registration frame that correspond to an expected aperture dimension as the plurality of registration frame aperture locations.

### Claim 2 (depends on 1)

The method of claim 1, further comprising gray scaling the image of the registration frame based on a threshold prior to performing the closing operation.

### Claim 3 (depends on 1)

The method of claim 1, wherein determining a pose of the registration frame comprises calculating an error between aperture representations of the model of the registration frame and the plurality of registration frame aperture locations within the image data.

### Claim 4 (depends on 1)

The method of claim 1, wherein an aperture of the plurality of apertures is cylindrical.

### Claim 5 (depends on 1)

The method of claim 1, wherein the determining a pose of the registration frame within the image space using the plurality of registration frame aperture locations within the image data comprises matching aperture representations of a model of the registration frame to the plurality of registration frame aperture locations within the image data.

### Claim 6

A method of performing, registration between a patient space and an image space, comprising:
imaging the patient space to generate image data of the image space, the image data comprising an image of a registration frame that is located in the patient space, the registration frame comprising a plurality of apertures;
identifying a plurality of registration frame aperture locations within the image data, each of the plurality of aperture locations within the image data respectively corresponding to one of the plurality of a res of the registration frame;
identifying a pose of the registration frame within the patient space using a tracking system for the patient space;
determining a pose of the registration frame within the image space using the plurality of registration frame aperture locations within the image data;
computing a transform between the patient space and the image space based on the pose of the registration frame within the patient space and the pose of the registration frame within the image space; and
optimizing the transform by checking the symmetry of a representation of an aperture of the plurality of the apertures of the registration frame according to an adjustment of the transform.

### Claim 7 (depends on 6)

The method of claim 6, wherein the pose of the registration frame within the patient space comprises a position of the registration frame within the patient space and an orientation of the registration frame within the patient space.

### Claim 8 (depends on 6)

The method of claim 6, wherein imaging the patient space comprises taking a computed tomography (CT) scan of the patient space.

### Claim 9

A method of performing registration between a patient space and an image space, comprising:
imaging the patient space to generate data of the image space, the image data comprising an image of a registration frame that is located in the patient space, the registration frame comprising a plurality of apertures;
identifying a plurality of registration frame aperture locations within the image data, each of the plurality of aperture locations within the image data respectively corresponding to one of the plurality of apertures of the registration frame;
identifying a pose of the registration frame within the patient space using a tracking system for the patient space;
determining a pose of the registration frame within the image space using the plurality of registration frame aperture locations within the image data;
computing a transform between the patient space and the image space based on the pose of the registration frame within the patient space and the pose of the registration frame within the image space; and
optimizing the transform by checking an alignment between a pair of plates of material of the registration frame taken according to an adjustment of the transform.

### Claim 10 (depends on 9)

The method of claim 9, wherein identifying a pose of a registration frame within the patient space using a tracking system for the patient space comprises tracking a tracking device of the registration frame with the tracking system.

### Claim 11 (depends on 7)

The method of claim 7, wherein the tracking device is an infrared (IR) reflector.

### Claim 12

A method of performing registration between a patient space and an image space, comprising:
imaging the patient space to generate image data of the image space, the image data comprising an image of a registration frame that is located in the patient space, the registration frame comprising a plurality of apertures;
identifying a plurality of registration frame aperture locations within the image data, each of the plurality of aperture locations within the image data respectively corresponding to one of the plurality of apertures of the registration frame;
identifying a pose of the registration frame within the patient space using a tracking system for the patient space;
determining a pose of the registration frame within the image space using the plurality of registration frame aperture locations within the image data;
computing a transform between the patient space and the image space based on se of the registration frame within the patient space and the pose of the registration frame within the image space;
generating a cropped image of the registration frame by applying the transform to the image data;
generating an optimized parameter set using a plurality of initial parameter sets and the cropped image of the registration frame; and
generating an optimized transform by applying the optimized parameter set to the transform.

### Claim 13 (depends on 12)

The method of claim 12, wherein generating the optimized parameter set using the plurality of initial parameter sets with the cropped image of the registration frame comprises:
determining a sub-plurality of the plurality of initial parameter sets that has the lowest associated cost using the cropped image of the registration frame;
optimizing between the sub-plurality of the plurality of initial parameter sets to generate the optimized parameter set.

### Claim 14 (depends on 13)

The method of claim 13, wherein determining the sub-plurality of the plurality of initial parameter sets that has the lowest associated cost using the cropped image of the registration frame is performed by, for each parameter set in the plurality of initial parameter sets:
creating an adjusted transform by adjusting the transform using the initial parameter set; and
comparing a first plate taken from the cropped image of the registration frame according to the adjusted transform to a second plate taken from the cropped image of the registration frame according to the adjusted transform to generate an associated cost for the initial parameter set.

### Claim 15 (depends on 13)

The method of claim 13, wherein determining the sub-plurality of the plurality of initial parameter sets that has the lowest associated cost using the cropped image of the registration frame is performed by, for each initial parameter set in the plurality of initial parameter sets:
creating an updated transform by adjusting the transform using the initial parameter set;
determining a modeled aperture location within the cropped image of the registration frame by applying one of the aperture representations within the model to the cropped image of the registration frame using the updated transform; and
comparing intensity values from voxels of a first portion of the modeled aperture location to intensity values from voxels of a second portion of the modeled aperture location to generate an associated cost for the initial parameter set.

### Claim 16 (depends on 13)

The method of claim 13, wherein optimizing between the sub-plurality of the plurality of parameter sets is performed using a Nelder-Mead optimization method.

### Claim 17 (depends on 12)

The method of claim 12, wherein at least one of the plurality of parameter sets is determined by:
generating a secondary transform by using one fewer of the plurality of registration frame aperture locations within the image data; and
calculating a measurement between the secondary transform and the transform.

### Claim 18

A computing apparatus comprising:
a processor; and
a memory storing instructions that, when executed by the processor, configure the apparatus to:
identify a pose of a registration frame within a patient space using a tracking system for the patient space, the registration frame comprising a plurality of apertures;
receive image data corresponding to an image space, the image data comprising an image of the registration frame;
identify a plurality of registration frame aperture locations within the image data, each of the plurality of aperture locations within the image data respectively corresponding to one of the plurality of apertures of the registration frame;
determine a pose of the registration frame within the image space using the plurality of registration frame aperture locations within the image data;
compute a transform between the patient space and the image space based on the pose of the registration frame within the patient space and the pose of the registration frame within the image space; and
optimize transform by checking the symmetry of a representation of an aperture of the plurality of the apertures of the registration frame according to an adjustment of the transform.

### Claim 19

A non-transitory computer-readable storage medium, the computer-readable storage medium including instructions that when executed by a computer, cause the computer to:
identify a pose of a registration frame within a patient space using a tracking system for the patient space, the registration frame comprising a plurality of apertures;
receive image data corresponding to an image space, the image data comprising an image of the registration frame;
identify a plurality of registration frame aperture locations within the image data using the image of the registration frame, each of the plurality of aperture locations within the image data respectively corresponding to one of the plurality of apertures of the registration frame;
determine a pose of the registration frame within the image space using the plurality of registration frame aperture locations within the image data;
compute a transform between the patient space and the image space based on the pose of the registration frame within the patient space and the pose of the registration frame within the image space;
generating a cropped image of the registration frame by applying the transform to the image data;
generating an optimize parameter set using a plurality of initial parameter sets and the cropped image of the registration frame; and
generating an optimized transform by applying the optimized parameter set to the transform.