AUGMENTED REALITY PATIENT POSITIONING USING AN ATLAS

Abstract

The disclosed method encompasses using an augmented reality device to blend in augmentation information including for example atlas information. The atlas information may be display separately from or in addition to a patient image (planning image). In order to display the atlas information in a proper position relative to the patient image, the two data sets are registered to one another. This registration can serve for generating a diversity of atlas-based image supplements, for example alternatively or additionally to the foregoing for displaying a segmentation of the patient image in the augmented reality image. The disclosed method is usable in a medical environment such as for surgery or radiotherapy.

Claims

1. A computer-implemented method for generating augmentation information relating to an image using output of an augmented reality device having an output unit for outputting the augmentation information comprising: acquiring patient medical image data which describes a patient image of an anatomical body part of a patient; acquiring geometric model data which describes a geometric model of at least part of the patient's body; registering the patient image with the geometric model; presenting augmentation information using the output unit of the augmented reality device based on a result of the registering the patient image with the geometric model and displaying as an overlay on a real image which is acquired using the augmented reality device and which describes the physical scene in the field of view of the augmented reality device or is acquired digitally and rendered on the output unit .

2. The method according to claim 1, wherein the geometric model data is acquired from atlas data describing the at least part of the patient's body or from a generic geometric model of the at least part of the patient's body, or from an individual synthetic template describing a synthesized geometric model of the at least part of the patient's body.

3. The method according to claim 1, wherein the geometric model comprises a model of at least part of the anatomical body part.

4. The method according to claim 1, wherein the patient image is registered with the geometric model by establishing a positional mapping between a patient image coordinate system in which image coordinates of the patient image are defined and a model coordinate system in which positions of the geometric model are defined by applying an image fusion algorithm to the patient image and the geometric model.

5. The method according to claim 1, wherein the augmentation information describes at least part of the geometric model of at least part of the anatomical body part.

6. The method according to claim 1, wherein the display or projector unit, respectively, of the augmented reality device is configured to output augmentation information as an overlay on the real image acquired using the augmented reality device.

7. The method according to claim 1, wherein a physical structure is at least part of a patient's body or a shape representative of at least part of a patient's body.

8. The method according to claim 7, wherein the physical structure is positionally tracked using a result of distance measurements by the distance measurement unit of the augmented reality device, and wherein registration between the at least one of at least part of the geometric model or at least part of the anatomical body part with the physical structure is used to determine, based on the result of the positional tracking, the position of at least part of the at least one of at least part of the geometric model or at least part of the anatomical body part in a coordinate system used by the augmented reality device to output image information describing the at least one of at least part of the geometric model or at least part of the anatomical body part.

9. The method according to claim 7, wherein the physical structure is positionally tracked using a result of distance measurements by the distance measurement unit of the augmented reality device, and wherein the registration between the at least one of at least part of the geometric model or at least part of the anatomical body part with the physical structure is used to determine, based on the result of the positional tracking, control data for controlling movement of at least part of the physical structure.

10. The method according to claim 1, further comprising the following features: acquiring medical procedure data describing an envisaged medical procedure to be carried out on the patient; acquiring device control data describing a control parameter of a medical device based on the medical procedure data; outputting augmentation information corresponding to the device control data by the augmented reality device.

11. A non-transitory computer readable medium comprising instructions which, when running on at least one processor of at least one computer, causes the at least one computer to: acquire patient medical image data which describes a patient image of an anatomical body part of the patient; acquire geometric model data which describes a geometric model of at least part of the patient's body; register the patient image with the geometric model; present augmentation information using an output unit of an augmented reality device based on a result of the registering the patient image with the geometric model and display as an overlay on a real image which is acquired using the augmented reality device and which describes the physical scene in the field of view of the augmented reality device or is acquired digitally and rendered on the output unit.

12. A medical system comprising: at least one computer; an augmented reality device having an output unit for outputting augmentation information; the at least one computer having at least one processor and associated memory, the memory having instructions causing the at least one processor to: acquire patient medical image data which describes a patient image of an anatomical body part of the patient; acquire geometric model data which describes a geometric model of at least part of the patient's body; register the patient image with the geometric model; present augmentation information using the output unit of the augmented reality device based on a result of the registering the patient image with the geometric model; and display as an overlay on a real image which is acquired using the augmented reality device and which describes the physical scene in the field of view of the augmented reality device or is acquired digitally and rendered on the output unit.

Description

DESCRIPTION OF THE FIGURES

[0058] In the following, the invention is described with reference to the appended figures which represent a specific embodiment of the invention. The scope of the invention is however not limited to the specific features disclosed in the context of the figures, wherein

[0059] FIG. 1 is a flow diagram illustrating the basic steps of the method according to the first aspect;

[0060] FIG. 2 illustrates a specific variant of the algorithm of FIG. 1;

[0061] FIGS. 3 to 4 illustrate four examples of augmentation information; and

[0062] FIG. 5 illustrates the setup of the system according to the sixth aspect.

[0063] FIG. 1 shows the basic flow of the method according to the first aspect, which starts with a step S11 of acquiring the patient medical image data, followed by a step S12 of acquiring the geometric model data. In subsequent step S13, the patient image is registered with the geometric model, followed by step S14 encompassing output of the augmentation information.

[0064] FIG. 2 illustrates a variation of the flow diagram of FIG. 1. Starting with step S21, the patient information (e.g. at least one of the patient medical image data or the patient administrative data or the medical procedure data) is acquired. In subsequent step S22, the geometric model data is acquired by acquiring atlas data which is then used to process (e.g. segment) the patient image. The registration between the atlas and the patient image is established in step S23 which encompasses creation of a virtual space with a patient surface model, a patient support unit (e.g. a table) and medical devices. The registration is then used in step S24 to determine an optimal position of the patient and the medical devices in the virtual space (i.e. in the coordinate system used by the augmented reality device to output image data). Corresponding augmentation information is displayed by the augmented reality device in step S25. The user may then modify in step S26 the position of the patient and/or the medical devices in the virtual space.

[0065] FIGS. 3 and 4 illustrate examples of augmentation information. For example, FIG. 3 shows a patient image describing the patient's head 1, combined with a real image of a patient table 2. This image is supplemented by augmentation information describing a head clamp 3. FIG. 4 illustrates augmentation information describing a positional range 5 defining possible desired positions of a medical device on a support rail 6. In this example, the medical device is embodied by an articulable arm 4.

[0066] FIG. 5 is a schematic illustration of the medical system 7 according to the sixth aspect. The system comprises an augmented reality device 8 having a digital processor and volatile memory (a computer 9) and an output unit 10 for outputting the augmentation information. The augmented reality device is operably coupled (indicated by the double arrow) to an external computer 11 which is configured to execute a program for conducting at least part of the method according to the first aspect.

[0067] The method according to the first aspect may be summarized in a non-limiting example as follows:

[0068] In a first step the patient's imaging data is analyzed. Medical images, anatomical atlas segmentation, indication-specific and procedure-specific information acquired from hospital computer systems is analyzed. The following sample attributes are deduced from the information:

[0069] From indication and procedure, the body part to be operated is identified (e.g. spine).

[0070] From patient administrative data, the length and weight of the patient is identified.

[0071] From imaging data, anatomical atlas and length/weight, a surface model of the patient is generated.

[0072] From imaging data, anatomical atlas, body part and procedure information, the exact location of the surgery is identified (e.g. a specific spine vertebra).

[0073] Based on the above information, a computer device augments a virtual world with:

[0074] A 3-dimensional model of the patient surface model that is positioned on top of an operation room (OR) table. Any real flat object would be used to simulate the OR table.

[0075] A recommended position for patient positioning is initially used in the display of the patient.

[0076] Recommended positions for medical devices and instruments that have image-specific positions outside of the patient's anatomy (e.g. Mayfield clamp or surgical robot). The positions could be displayed through virtual devices or virtual markers in an augmentation device. The virtual markers define the desired position of the device and/or desired modifications to the device (e.g. Mayfield clamp tightness).

[0077] Recommended positions for medical devices and planning objects that have image-specific positions inside the patient's anatomy (e.g. screws in spine surgery).

[0078] The recommended positions are marked to indicate most recommended, medium recommended and least recommended areas (e.g. through color or pattern marks).

[0079] In a further step the surgeon, a nurse or any other OR staff uses augmented reality glasses (or any other augmentation device) to view the virtual world described above. The world is blended into (augmented) the room currently occupied by the person wearing the glasses. This could be inside the OR or in another room, e.g. surgeon's office.

[0080] In a final step, through interaction with the augmentation device, it is possible to modify the position of patient and devices as well as remove, add or correct patient anatomy features and devices.

[0081] Alternatively, a real device could be used to modify the patient position. It could be a scale representation of a patient, e.g. a small-scale model of a patient that consists of modifiable arms, legs and head. The different-scale representation of the patient anatomy is seen by the sensors/cameras on the augmentation device and is used to modify the position of the virtual patient on the table.

[0082] An individual wearing or interacting with the augmentation device can see and interact with recommended positions of patient and devices in a much more intuitive way than without augmentation. The visualization can augment the OR where the procedure will take place or, alternatively, augment another location different from the OR.

[0083] This provides the benefit of image-based automatic suggestions for OR setup. They can be used by OR staff to prepare for surgery. The innovation also enables surgeons to in detail plan and approve OR setups independently of their physical location.