WO2025214601A1

WO2025214601A1 - Markerless registration with mixed reality and face pose estimation

Info

Publication number: WO2025214601A1
Application number: PCT/EP2024/059800
Authority: WO
Inventors: Octavio RODRÍGUEZ; Laura SCHMISS
Original assignee: Brainlab SE
Current assignee: Brainlab SE
Priority date: 2024-04-11
Filing date: 2024-04-11
Publication date: 2025-10-16
Anticipated expiration: 2026-10-11

Abstract

Disclosed is a computer-implemented method of establishing a spatial transformation between a two-dimensional digital camera image and a three-dimensional digital model of an anatomical body part. The disclosed method encompasses transformation of an RGB camera image of an anatomical body part of patient taken with a head-mounted device into a point cloud, scaling the point cloud according to a relation of distances between anatomical landmarks in the camera image and/or the point cloud to distances between corresponding anatomical landmarks in a three-dimensional planning image of the anatomical body part, and establishing a spatial transformation between the scaled point cloud and the planning image. This spatial transformation can be used for augmenting an image or real scene viewed by a user of the head-mounted device with the planning image as an overlay on the position of the anatomical body part viewed by the user. Thereby, user guidance is improved.

Description

MARKERLESS REGISTRATION WITH MIXED REALITY AND FACE POSE ESTIMATION

FIELD OF THE INVENTION

The present invention relates to a computer-implemented method of establishing a spatial transformation between a two-dimensional digital camera image and a three- dimensional digital model of an anatomical body part, a corresponding computer program, a computer-readable storage medium storing such a program and a computer executing the program, as well as a medical system comprising an electronic data storage device and the aforementioned computer.

TECHNICAL BACKGROUND

Certain medical interventions (e.g. extra ventricular drain placement) ca be made faster and simpler without sacrificing accuracy by determining the position and orientation of a patient’s head in order to overlay three-dimensional planning data on top of an actual image of the patient.

A sensor, e.g. depth camera, arrays of colour image cameras or an inertial measurement unit, of a mixed reality device are used to keep track of the position of the device in space. We want to use those sensors to estimate the position and orientation of a patient’s face, without the need of additional equipment or markers. Running an artificial intelligence algorithm on an RGB (red, green, and blue) images, the images can be analysed to detect whether human faces are present and to estimate the orientation of the face (see Fig. 3).

Previous attempts to overlay patient data using MR devices have: • yielded unsatisfactory results - considerable deviations between 3D data and the patient;

• required manual positioning - which requires time and skill; and

• required special markers - which require planning ahead, additional scans, and time.

This is largely due to the fact that these approaches depend on the depth sensors on the mixed reality devices, and therefore on their accuracy.

The present invention has the object of providing improved image-based means for determining the position and orientation of an anatomical body part.

Aspects of the present invention, examples and exemplary steps and their embodiments are disclosed in the following. Different exemplary features of the invention can be combined in accordance with the invention wherever technically expedient and feasible.

EXEMPLARY SHORT DESCRIPTION OF THE INVENTION

In the following, a short description of the specific features of the present invention is given which shall not be understood to limit the invention only to the features or a combination of the features described in this section.

The disclosed method encompasses transformation of an RGB camera image of ana anatomical body part of patient taken with a head-mounted device into a point cloud, scaling the point cloud according to a relation of distances between anatomical landmarks in the camera image and/or the point cloud to distances between corresponding anatomical landmarks in a three-dimensional planning image of the anatomical body part, and establishing a spatial transformation between the scaled point cloud and the planning image. This spatial transformation can be used for augmenting an image or real scene viewed by a user of the head-mounted device with the planning image as an overlay on the position of the anatomical body part viewed by the user. Thereby, user guidance is improved. GENERAL DESCRIPTION OF THE INVENTION

In this section, a description of the general features of the present invention is given for example by referring to possible embodiments of the invention.

In general, the invention reaches the aforementioned object by providing, in a first aspect, a computer-implemented method of establishing a spatial transformation between a two-dimensional digital camera image and a three-dimensional digital model of an anatomical body part. The method is for example a medical method. The method comprises executing, on at least one processor of at least one computer (for example at least one computer being part of a navigation system), the following exemplary steps which are executed by the at least one processor.

In a (for example first) exemplary step, camera image data is acquired which describes a two-dimensional digital camera image of the anatomical body part. The anatomical body part belongs for example to a patient and for example is or comprises at least part of the face. However, the anatomical body part can alternatively or additionally consist of or comprises other part of the patient’s body, for example the spine or the pelvis. For example, the two-dimensional camera image has been taken with an imaging device, for example a two-dimensional RGB camera, included in a mixed reality device such as a mixed reality/augmented reality device, for example headmounted device such as mixed or augmented reality goggles.

In a (for example second) exemplary step, patient image data is acquired which describes a three-dimensional digital patient image (for example, a planning image) of the anatomical body part. For example, the patient image data has been generated by applying a tomographic imaging modality such as computed x-ray tomography, magnetic resonance tomography or ultrasound tomography to the anatomical body part.

In a (for example third) exemplary step, body part position data is determined based on the camera image data. The body part position data describes a position, for example in the camera image, of the image representation of the anatomical body part in the camera image. For example, the body part position data is determined by inputting the camera image data into an algorithm such as a trainable algorithm, i.e. learning algorithm, for example an artificial intelligence algorithm such as a DNN (distributed neural network) algorithm, configured, i.e. trained, to determine the position, in the camera image, of the image representation of the anatomical body part. Instead of a DNN, a convolutional neural network (e.g. Haar Cascade Classifiers as disclosed at https://docs.opencv.Org/4.x/d2/d99/tutorialjs_face_detection.html) can be used to determine the position, in the camera image, of the image representation of the anatomical body part. A computer graphics approach for to determining the position, in the camera image, of the image representation of the anatomical body part without neural networks) can be based on Eigenfaces (see https://en.wikipedia.org/wiki/Eigenface).

In a (for example fourth) exemplary step, point cloud data is determined based on the body part position data. The point cloud data describes a three-dimensional point cloud representation of the image representation of the anatomical body part and includes landmark point data describing the positions of points of the point cloud representing predetermined parts of the anatomical body part. For example, the predetermined parts are anatomical landmarks. The point cloud data is determined for example by applying a hierarchical representation network (HRN) to the body part position data, i.e. by inputting the body part position data to an artificial intelligence algorithm such as an HRN algorithm. For example, the point cloud data is determined by inputting the camera image data into an algorithm such as a trainable algorithm, i.e. learning algorithm, for example an artificial intelligence algorithm (such as the aforementioned HRN algorithm), configured, i.e. trained, to transform the image representation of the anatomical body part in the camera image into the point cloud and to determine the positions of the points of the point cloud representing the predetermined parts of the anatomical body part. Details of the HRN algorithm are available at https://github.com/youngLBW/HRN and have been published in the paper by Lei, B. et al., A Hierarchical Representation Network for Accurate and Detailed Face Reconstruction from In-The-Wild Images, CVPR2023, https://doi.Org/10.48550/arXiv.2302.14434.

In a (for example fifth) exemplary step, landmark position data is determined based on the patient image data, wherein the landmark position data describes the positions of the image representations of the predetermined parts in the patient image. For example, atlas data is acquired which describes a three-dimensional digital model of the anatomical body part and an identification of parts of the model corresponding to the predetermined parts of the anatomical body part, and the landmark position data is determined based on the patient image data and the atlas data, for example by matching, i.e. registering, the patient image data with the atlas data. Alternatively, the landmark position data is determined by an artificial algorithm configured to determine the positions of the image representations of the predetermined parts in the patient image, for example the aforementioned artificial intelligence, for example HRN, algorithm. In a further alternative, a specific determination algorithm which has been developed for this purpose may be used.

In a (for example sixth) exemplary step, landmark matching data is determined based on the point cloud data, for example based on the landmark point data, and the landmark position data. The landmark matching data describes a correspondence between the image representations of the predetermined parts in the patient image and each of the points of the point cloud representing the predetermined parts. In other word, the landmark matching data describes an assignment of points in the point cloud and the image representations of the predetermined parts in the patient image.

In a (for example seventh) exemplary step, landmark scaling data is determined based on the landmark matching data. The landmark scaling data describes a relation such as a scaling or scaling factor or a numeric value of the a scaling or a scaling factor between distances between the positions of each a, for example one or exactly one, pair of points of the point cloud representing predetermined parts and distances between the positions of each a, for example one or exactly one, pair of anatomically corresponding ones of the image representations of the predetermined parts in the patient image, i.e. to the points of the point cloud representing the corresponding predetermined parts.

In a (for example eighth) exemplary step, scaled point cloud data is determined based on the landmark scaling data and the point cloud data, wherein the scaled point cloud data describes a scaling of the point cloud. The scaling is the result of applying the relation described by the landmark scaling data, for example the scaling or scaling factor, to the point cloud described by the point cloud data.

In a (for example ninth) exemplary step, transformation data is determined based on the scaled point cloud data and the landmark position data. The transformation data describes a positional transformation, for example a matching or registration or mapping, between the positions of the points representing the predetermined parts of the anatomical body part in the scaling of the point cloud, i.e. the scaled point cloud, and the positions of the image representations of the predetermined parts of the anatomical body part in the patient image.

In an example of the method according to the first aspect, patient image transformation data is determined based on the patient image data and the transformation data. For example, the patient image transformation data describes a positional orientation of the patient image to the position of the anatomical body part. For example, the patient image transformation data describes the result of an application of the positional transformation to the patient image, in a reference system of an imaging device used to generate the camera image data. In this example, augmentation image data is optionally determined based on the camera image data and the patient image transformation data. The augmentation image data describes an image augmentation of the representation of the anatomical body part in the patient image and the camera image and/or at least part of a field of view of an imaging device used to generate the camera image (for example an overlay of the representation of the anatomical body part in the patient image onto the camera image and/or at least part of a field of view of an imaging device used to generate the camera image). For example, camera tracking data is acquired which describes the position of the imaging device used to acquire the camera image data, and the augmentation image data is determined based on the camera tracking data and the transformation data. For example, the image augmentation is displayed on a display device, wherein the display device and an imaging device used to generate the camera image data are integrated into the same device, for example a mixed reality device. For example, the augmentation image data is used to determine the position of a situs on the anatomical body part for performing a medical procedure, for example a craniotomy. In an example of the method according to the first aspect, scaling factor data is determined based on the landmark scaling data. The landmark scaling data describes a plurality of, for example at least two, relations between the distances between the positions of each a, for example one or exactly one pair of points of the point cloud representing predetermined parts and distances between the positions of each a, for example one or exactly one, pair of anatomically corresponding ones of the image representations of the predetermined parts in the patient image (i.e. the image representations of the predetermined parts in the patient image corresponding to the points of the point cloud representing predetermined parts), and the scaling factor data describes a mean scaling factor, for example a value of a mean scaling factor, determined from the relations described by the landmark scaling data. The scaled point cloud data is determined based on the scaling factor data.

In an example of the method according to the first aspect, control data for controlling a medical device is determined based on the augmentation image data. For example, the control data can be used for controlling a medical device such a robot, e.g. surgical robot, to attain a predetermined position relative to the anatomical body part or conduct a medical procedure on the anatomical body part. For example, the method according to the first aspect encompasses executing the control data, i.e. issuing the control data to the medical device for controlling the medical device, i.e. the operation of the medical device.

In a second aspect, the invention is directed to a computer program comprising instructions which, when the program is executed by at least one computer, causes the at least one computer to carry out method according to the first aspect. The invention may alternatively or additionally relate to a (physical, for example electrical, for example technically generated) signal wave, for example a digital signal wave, such as an electromagnetic carrier wave carrying information which represents the program, for example the aforementioned program, which for example comprises code means which are adapted to perform any or all of the steps of the method according to the first aspect. The signal wave is in one example a data carrier signal carrying the aforementioned computer program. A computer program stored on a disc is a data file, and when the file is read out and transmitted it becomes a data stream for example in the form of a (physical, for example electrical, for example technically generated) signal. The signal can be implemented as the signal wave, for example as the electromagnetic carrier wave which is described herein. For example, the signal, for example the signal wave is constituted to be transmitted via a computer network, for example LAN, WLAN, WAN, mobile network, for example the internet. For example, the signal, for example the signal wave, is constituted to be transmitted by optic or acoustic data transmission. The invention according to the second aspect therefore may alternatively or additionally relate to a data stream representative of the aforementioned program, i.e. comprising the program.

In a third aspect, the invention is directed to a computer-readable storage medium on which the program according to the second aspect is stored. The program storage medium is for example non-transitory.

In a fourth aspect, the invention is directed to at least one computer (for example, a computer), comprising at least one processor (for example, a processor), wherein the program according to the second aspect is executed by the processor, or wherein the at least one computer comprises the computer-readable storage medium according to the third aspect.

In a fifth aspect, the invention is directed to a medical system, comprising: a) the at least one computer according to the fourth aspect, wherein the program comprises instructions which, when the program is executed by a computer, cause the computer to carry out the method according to the first aspect, for example as far as the method according to the first aspect comprises acquisition of the camera tracking data and determination of the augmentation image data; b) at least one electronic data storage device storing at least the patient image data; and c) a mixed reality device for acquiring the camera image data and displaying the augmentation image data, wherein the at least one computer is operably coupled to the at least one electronic data storage device for acquiring, from the at least one data storage device, at least the patient image data, and wherein the at least one computer is operably coupled to or part of the mixed reality device for displaying the image augmentation on a display of the mixed reality device on the basis of the augmentation image data and the camera tracking data.

For example, the disclosed method is not a method for treatment of the human or animal body by surgery or therapy. For example, the invention does not involve or in particular comprise or encompass an invasive step which would represent a substantial physical interference with the body requiring professional medical expertise to be carried out and entailing a substantial health risk even when carried out with the required professional care and expertise. More particularly, the invention does not involve or in particular comprise or encompass any surgical or therapeutic activity. The invention is instead directed as applicable to processing image data. For this reason alone, no surgical or therapeutic activity and in particular no surgical or therapeutic step is necessitated or implied by carrying out the invention.

DEFINITIONS

In this section, definitions for specific terminology used in this disclosure are offered which also form part of the present disclosure.

The method in accordance with the invention is for example a computer-implemented method. For example, all the steps or merely some of the steps (i.e. less than the total number of steps) of the method in accordance with the invention can be executed by a computer (for example, at least one computer). An embodiment of the computer implemented method is a use of the computer for performing a data processing method. An embodiment of the computer implemented method is a method concerning the operation of the computer such that the computer is operated to perform one, more or all steps of the method.

The computer for example comprises at least one processor and for example at least one memory in order to (technically) process the data, for example electronically and/or optically. The processor being for example made of a substance or composition which is a semiconductor, for example at least partly n- and/or p-doped semiconductor, for example at least one of II-, III- IV-, V-, Vl-sem iconductor material, for example (doped) silicon and/or gallium arsenide. The calculating or determining steps described are for example performed by a computer. Determining steps or calculating steps are for example steps of determining data within the framework of the technical method, for example within the framework of a program. A computer is for example any kind of data processing device, for example electronic data processing device. A computer can be a device which is generally thought of as such, for example desktop PCs, notebooks, netbooks, etc., but can also be any programmable apparatus, such as for example a mobile phone or an embedded processor. A computer can for example comprise a system (network) of "sub-computers", wherein each sub-computer represents a computer in its own right. The term "computer" includes a cloud computer, for example a cloud server. The term computer includes a server resource. The term "cloud computer" includes a cloud computer system which for example comprises a system of at least one cloud computer and for example a plurality of operatively interconnected cloud computers such as a server farm. Such a cloud computer is preferably connected to a wide area network such as the world wide web (WWW) and located in a so-called cloud of computers which are all connected to the world wide web. Such an infrastructure is used for "cloud computing", which describes computation, software, data access and storage services which do not require the end user to know the physical location and/or configuration of the computer delivering a specific service. For example, the term "cloud" is used in this respect as a metaphor for the Internet (world wide web). For example, the cloud provides computing infrastructure as a service (laaS). The cloud computer can function as a virtual host for an operating system and/or data processing application which is used to execute the method of the invention. The cloud computer is for example an elastic compute cloud (EC2) as provided by Amazon Web Services™. A computer for example comprises interfaces in order to receive or output data and/or perform an analogue-to-digital conversion. The data are for example data which represent physical properties and/or which are generated from technical signals. The technical signals are for example generated by means of (technical) detection devices (such as for example devices for detecting marker devices) and/or (technical) analytical devices (such as for example devices for performing (medical) imaging methods), wherein the technical signals are for example electrical or optical signals. The technical signals for example represent the data received or outputted by the computer. The computer is preferably operatively coupled to a display device which allows information outputted by the computer to be displayed, for example to a user. One example of a display device is a virtual reality device or an augmented reality device (also referred to as virtual reality glasses or augmented reality glasses) which can be used as "goggles" for navigating. A specific example of such augmented reality glasses is Google Glass (a trademark of Google, Inc.). An augmented reality device or a virtual reality device can be used both to input information into the computer by user interaction and to display information outputted by the computer. Another example of a display device would be a standard computer monitor comprising for example a liquid crystal display operatively coupled to the computer for receiving display control data from the computer for generating signals used to display image information content on the display device. A specific embodiment of such a computer monitor is a digital lightbox. An example of such a digital lightbox is Buzz®, a product of Brainlab AG. The monitor may also be the monitor of a portable, for example handheld, device such as a smart phone or personal digital assistant or digital media player.

The invention also relates to a computer program comprising instructions which, when on the program is executed by a computer, cause the computer to carry out the method or methods, for example, the steps of the method or methods, described herein and/or to a computer-readable storage medium (for example, a non-transitory computer- readable storage medium) on which the program is stored and/or to a computer comprising said program storage medium and/or to a (physical, for example electrical, for example technically generated) signal wave, for example a digital signal wave, such as an electromagnetic carrier wave carrying information which represents the program, for example the aforementioned program, which for example comprises code means which are adapted to perform any or all of the method steps described herein. The signal wave is in one example a data carrier signal carrying the aforementioned computer program. The invention also relates to a computer comprising at least one processor and/or the aforementioned computer-readable storage medium and for example a memory, wherein the program is executed by the processor.

Within the framework of the invention, computer program elements can be embodied by hardware and/or software (this includes firmware, resident software, micro-code, etc.). Within the framework of the invention, computer program elements can take the form of a computer program product which can be embodied by a computer-usable, for example computer-readable data storage medium comprising computer-usable, for example computer-readable program instructions, "code" or a "computer program" embodied in said data storage medium for use on or in connection with the instructionexecuting system. Such a system can be a computer; a computer can be a data processing device comprising means for executing the computer program elements and/or the program in accordance with the invention, for example a data processing device comprising a digital processor (central processing unit or CPU) which executes the computer program elements, and optionally a volatile memory (for example a random access memory or RAM) for storing data used for and/or produced by executing the computer program elements. Within the framework of the present invention, a computer-usable, for example computer-readable data storage medium can be any data storage medium which can include, store, communicate, propagate or transport the program for use on or in connection with the instruction-executing system, apparatus or device. The computer-usable, for example computer-readable data storage medium can for example be, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared or semiconductor system, apparatus or device or a medium of propagation such as for example the Internet. The computer-usable or computer-readable data storage medium could even for example be paper or another suitable medium onto which the program is printed, since the program could be electronically captured, for example by optically scanning the paper or other suitable medium, and then compiled, interpreted or otherwise processed in a suitable manner. The data storage medium is preferably a non-volatile data storage medium. The computer program product and any software and/or hardware described here form the various means for performing the functions of the invention in the example embodiments. The computer and/or data processing device can for example include a guidance information device which includes means for outputting guidance information. The guidance information can be outputted, for example to a user, visually by a visual indicating means (for example, a monitor and/or a lamp) and/or acoustically by an acoustic indicating means (for example, a loudspeaker and/or a digital speech output device) and/or tactilely by a tactile indicating means (for example, a vibrating element or a vibration element incorporated into an instrument). For the purpose of this document, a computer is a technical computer which for example comprises technical, for example tangible components, for example mechanical and/or electronic components. Any device mentioned as such in this document is a technical and for example tangible device.

The expression "acquiring data" for example encompasses (within the framework of a computer implemented method) the scenario in which the data are determined by the computer implemented method or program. Determining data for example encompasses measuring physical quantities and transforming the measured values into data, for example digital data, and/or computing (and e.g. outputting) the data by means of a computer and for example within the framework of the method in accordance with the invention. A step of “determining” as described herein for example comprises or consists of issuing a command to perform the determination described herein. For example, the step comprises or consists of issuing a command to cause a computer, for example a remote computer, for example a remote server, for example in the cloud, to perform the determination. Alternatively or additionally, a step of “determination” as described herein for example comprises or consists of receiving the data resulting from the determination described herein, for example receiving the resulting data from the remote computer, for example from that remote computer which has been caused to perform the determination. The meaning of "acquiring data" also for example encompasses the scenario in which the data are received or retrieved by (e.g. input to) the computer implemented method or program, for example from another program, a previous method step or a data storage medium, for example for further processing by the computer implemented method or program. Generation of the data to be acquired may but need not be part of the method in accordance with the invention. The expression "acquiring data" can therefore also for example mean waiting to receive data and/or receiving the data. The received data can for example be inputted via an interface. The expression "acquiring data" can also mean that the computer implemented method or program performs steps in order to (actively) receive or retrieve the data from a data source, for instance a data storage medium (such as for example a ROM, RAM, database, hard drive, etc.), or via the interface (for instance, from another computer or a network). The data acquired by the disclosed method or device, respectively, may be acquired from a database located in a data storage device which is operably to a computer for data transfer between the database and the computer, for example from the database to the computer. The computer acquires the data for use as an input for steps of determining data. The determined data can be output again to the same or another database to be stored for later use. The database or database used for implementing the disclosed method can be located on network data storage device or a network server (for example, a cloud data storage device or a cloud server) or a local data storage device (such as a mass storage device operably connected to at least one computer executing the disclosed method). The data can be made "ready for use" by performing an additional step before the acquiring step. In accordance with this additional step, the data are generated in order to be acquired. The data are for example detected or captured (for example by an analytical device). Alternatively or additionally, the data are inputted in accordance with the additional step, for instance via interfaces. The data generated can for example be inputted (for instance into the computer). In accordance with the additional step (which precedes the acquiring step), the data can also be provided by performing the additional step of storing the data in a data storage medium (such as for example a ROM, RAM, CD and/or hard drive), such that they are ready for use within the framework of the method or program in accordance with the invention. The step of "acquiring data" can therefore also involve commanding a device to obtain and/or provide the data to be acquired. In particular, the acquiring step does not involve an invasive step which would represent a substantial physical interference with the body, requiring professional medical expertise to be carried out and entailing a substantial health risk even when carried out with the required professional care and expertise. In particular, the step of acquiring data, for example determining data, does not involve a surgical step and in particular does not involve a step of treating a human or animal body using surgery or therapy. In order to distinguish the different data used by the present method, the data are denoted (i.e. referred to) as "XY data" and the like and are defined in terms of the information which they describe, which is then preferably referred to as "XY information" and the like.

Preferably, atlas data is acquired which describes (for example defines, more particularly represents and/or is) a general three-dimensional shape of the anatomical body part. The atlas data therefore represents an atlas of the anatomical body part. An atlas typically consists of a plurality of generic models of objects, wherein the generic models of the objects together form a complex structure. For example, the atlas constitutes a statistical model of a patient’s body (for example, a part of the body) which has been generated from anatomic information gathered from a plurality of human bodies, for example from medical image data containing images of such human bodies. In principle, the atlas data therefore represents the result of a statistical analysis of such medical image data for a plurality of human bodies. This result can be output as an image - the atlas data therefore contains or is comparable to medical image data. Such a comparison can be carried out for example by applying an image fusion algorithm which conducts an image fusion between the atlas data and the medical image data. The result of the comparison can be a measure of similarity between the atlas data and the medical image data. The atlas data comprises image information (for example, positional image information) which can be matched (for example by applying an elastic or rigid image fusion algorithm) for example to image information (for example, positional image information) contained in medical image data so as to for example compare the atlas data to the medical image data in order to determine the position of anatomical structures in the medical image data which correspond to anatomical structures defined by the atlas data.

The human bodies, the anatomy of which serves as an input for generating the atlas data, advantageously share a common feature such as at least one of gender, age, ethnicity, body measurements (e.g. size and/or mass) and pathologic state. The anatomic information describes for example the anatomy of the human bodies and is extracted for example from medical image information about the human bodies. The atlas of a femur, for example, can comprise the head, the neck, the body, the greater trochanter, the lesser trochanter and the lower extremity as objects which together make up the complete structure. The atlas of a brain, for example, can comprise the telencephalon, the cerebellum, the diencephalon, the pons, the mesencephalon and the medulla as the objects which together make up the complex structure. One application of such an atlas is in the segmentation of medical images, in which the atlas is matched to medical image data, and the image data are compared with the matched atlas in order to assign a point (a pixel or voxel) of the image data to an object of the matched atlas, thereby segmenting the image data into objects.

For example, the atlas data includes information of the anatomical body part. This information is for example at least one of patient-specific, non-patient-specific, indication-specific or non-indication-specific. The atlas data therefore describes for example at least one of a patient-specific, non-patient-specific, indication-specific or non-indication-specific atlas. For example, the atlas data includes movement information indicating a degree of freedom of movement of the anatomical body part with respect to a given reference (e.g. another anatomical body part). For example, the atlas is a multimodal atlas which defines atlas information for a plurality of (i.e. at least two) imaging modalities and contains a mapping between the atlas information in different imaging modalities (for example, a mapping between all of the modalities) so that the atlas can be used for transforming medical image information from its image depiction in a first imaging modality into its image depiction in a second imaging modality which is different from the first imaging modality or to compare (for example, match or register) images of different imaging modality with one another.

In the field of medicine, imaging methods (also called imaging modalities and/or medical imaging modalities) are used to generate image data (for example, two- dimensional or three-dimensional image data) of anatomical structures (such as soft tissues, bones, organs, etc.) of the human body. The term "medical imaging methods" is understood to mean (advantageously apparatus-based) imaging methods (for example so-called medical imaging modalities and/or radiological imaging methods) such as for instance computed tomography (CT) and cone beam computed tomography (CBCT, such as volumetric CBCT), X-ray tomography, magnetic resonance tomography (MRT or MRI), conventional X-ray, sonography and/or ultrasound examinations, and positron emission tomography. For example, the medical imaging methods are performed by the analytical devices. Examples for medical imaging modalities applied by medical imaging methods are: X-ray radiography, magnetic resonance imaging, medical ultrasonography or ultrasound, endoscopy, elastography, tactile imaging, thermography, medical photography and nuclear medicine functional imaging techniques as positron emission tomography (PET) and Single-photon emission computed tomography (SPECT), as mentioned by Wikipedia. The image data thus generated is also termed “medical imaging data”. Analytical devices for example are used to generate the image data in apparatusbased imaging methods. The imaging methods are for example used for medical diagnostics, to analyse the anatomical body in order to generate images which are described by the image data. The imaging methods are also for example used to detect pathological changes in the human body. However, some of the changes in the anatomical structure, such as the pathological changes in the structures (tissue), may not be detectable and for example may not be visible in the images generated by the imaging methods. A tumour represents an example of a change in an anatomical structure. If the tumour grows, it may then be said to represent an expanded anatomical structure. This expanded anatomical structure may not be detectable; for example, only a part of the expanded anatomical structure may be detectable. Primary/high- grade brain tumours are for example usually visible on MRI scans when contrast agents are used to infiltrate the tumour. MRI scans represent an example of an imaging method. In the case of MRI scans of such brain tumours, the signal enhancement in the MRI images (due to the contrast agents infiltrating the tumour) is considered to represent the solid tumour mass. Thus, the tumour is detectable and for example discernible in the image generated by the imaging method. In addition to these tumours, referred to as "enhancing" tumours, it is thought that approximately 10% of brain tumours are not discernible on a scan and are for example not visible to a user looking at the images generated by the imaging method.

Mapping describes a transformation (for example, linear transformation) of an element (for example, a pixel or voxel), for example the position of an element, of a first data set in a first coordinate system to an element (for example, a pixel or voxel), for example the position of an element, of a second data set in a second coordinate system (which may have a basis which is different from the basis of the first coordinate system). In one embodiment, the mapping is determined by comparing (for example, matching) the color values (for example grey values) of the respective elements by means of an elastic or rigid fusion algorithm. The mapping is embodied for example by a transformation matrix (such as a matrix defining an affine transformation).

Image fusion can be elastic image fusion or rigid image fusion. In the case of rigid image fusion, the relative position between the pixels of a 2D image and/or voxels of a 3D image is fixed, while in the case of elastic image fusion, the relative positions are allowed to change.

In this application, the term "image morphing" is also used as an alternative to the term "elastic image fusion", but with the same meaning. Elastic fusion transformations (for example, elastic image fusion transformations) are for example designed to enable a seamless transition from one dataset (for example a first dataset such as for example a first image) to another dataset (for example a second dataset such as for example a second image). The transformation is for example designed such that one of the first and second datasets (images) is deformed, for example in such a way that corresponding structures (for example, corresponding image elements) are arranged at the same position as in the other of the first and second images. The deformed (transformed) image which is transformed from one of the first and second images is for example as similar as possible to the other of the first and second images. Preferably, (numerical) optimisation algorithms are applied in order to find the transformation which results in an optimum degree of similarity. The degree of similarity is preferably measured by way of a measure of similarity (also referred to in the following as a "similarity measure"). The parameters of the optimisation algorithm are for example vectors of a deformation field. These vectors are determined by the optimisation algorithm in such a way as to result in an optimum degree of similarity. Thus, the optimum degree of similarity represents a condition, for example a constraint, for the optimisation algorithm. The bases of the vectors lie for example at voxel positions of one of the first and second images which is to be transformed, and the tips of the vectors lie at the corresponding voxel positions in the transformed image. A plurality of these vectors is preferably provided, for instance more than twenty or a hundred or a thousand or ten thousand, etc. Preferably, there are (other) constraints on the transformation (deformation), for example in order to avoid pathological deformations (for instance, all the voxels being shifted to the same position by the transformation). These constraints include for example the constraint that the transformation is regular, which for example means that a Jacobian determinant calculated from a matrix of the deformation field (for example, the vector field) is larger than zero, and also the constraint that the transformed (deformed) image is not self-intersecting and for example that the transformed (deformed) image does not comprise faults and/or ruptures. The constraints include for example the constraint that if a regular grid is transformed simultaneously with the image and in a corresponding manner, the grid is not allowed to interfold at any of its locations. The optimising problem is for example solved iteratively, for example by means of an optimisation algorithm which is for example a first-order optimisation algorithm, such as a gradient descent algorithm. Other examples of optimisation algorithms include optimisation algorithms which do not use derivations, such as the downhill simplex algorithm, or algorithms which use higher-order derivatives such as Newton-like algorithms. The optimisation algorithm preferably performs a local optimisation. If there is a plurality of local optima, global algorithms such as simulated annealing or generic algorithms can be used. In the case of linear optimisation problems, the simplex method can for instance be used.

In the steps of the optimisation algorithms, the voxels are for example shifted by a magnitude in a direction such that the degree of similarity is increased. This magnitude is preferably less than a predefined limit, for instance less than one tenth or one hundredth or one thousandth of the diameter of the image, and for example about equal to or less than the distance between neighbouring voxels. Large deformations can be implemented, for example due to a high number of (iteration) steps.

The determined elastic fusion transformation can for example be used to determine a degree of similarity (or similarity measure, see above) between the first and second datasets (first and second images). To this end, the deviation between the elastic fusion transformation and an identity transformation is determined. The degree of deviation can for instance be calculated by determining the difference between the determinant of the elastic fusion transformation and the identity transformation. The higher the deviation, the lower the similarity, hence the degree of deviation can be used to determine a measure of similarity.

A measure of similarity can for example be determined on the basis of a determined correlation between the first and second datasets.

BRIEF DESCRIPTION OF THE DRAWINGS

In the following, the invention is described with reference to the appended figures which give background explanations and represent specific embodiments of the invention. The scope of the invention is however not limited to the specific features disclosed in the context of the figures, wherein Fig. 1 illustrates the basic steps of the method according to the first aspect;

Fig. 2 shows an embodiment of the present invention, specifically the method according to the first aspect;

Fig. 3 illustrates face and landmark detection in the camera image;

Fig. 4 shows the principle of face reconstruction from the camera image;

Fig. 5 illustrates a conversion of a surface of the three-dimensional morphed model of the face into a three-dimensional point cloud;

Fig, 6 illustrates an alignment of the point cloud to the patient image; and

Fig. 7 is a schematic illustration of the system according to the fifth aspect.

DESCRIPTION OF EMBODIMENTS

Fig. 1 illustrates the basic steps of the method according to the first aspect, in which step S11 encompasses acquisition of the camera image data and step S12 encompasses acquisition of the patient image data. Subsequent step S13 encompasses determination of the body part position data, followed by step S14 which is directed to determining the point cloud data: Step S15 determines the landmark position data, and step S16 determines the landmark matching data. The landmark scaling data is determined in step S17, which is followed by determining the scaled point cloud data in step S18 and the transformation data in step S19.

Fig. 2 illustrates an embodiment of the present invention that includes all essential features of the invention. In this embodiment, the entire data processing which is part of the method according to the first aspect is performed by a computer 2. Reference sign 1 denotes the input of data acquired by the method according to the first aspect into the computer 2 and reference sign 3 denotes the output of data determined by the method according to the first aspect.

Fig. 3 shows extraction of the positions of landmarks 9 shown in a camera image 8 of a patient’s face by matching the camera image 8 with a model 10 of a face including a model of the positions of landmarks 11 . The camera image 8 has been taken with a camera included in a head-mounted device 12 which uses a camera coordinate system 13 for defining positions in the camera coordinate system. A registration (R, t) is established between the camera coordinate system and a world coordinate system in which position in the model 10 are defined.

The following steps outline the main idea of the registration process using an MR device:

1 . The mixed reality device loads DICOM data in the form of a 3three-dimensional reconstruction of the patient’s head and displays it.

2. The mixed reality device takes images with its RGB camera. Each image is analyzed by a DNN algorithm, which will attempt to detect whether there is a face in the image. If so: a. The region of the image where the face is, is forwarded to an HRN algorithm that estimates the position and orientation of the face: i. As shown in Fig. 4, a three-dimensional face reconstruction 16 from a single two-dimensional RGB image is determined based on a three-dimensional morphable model (3DMM) 10. The 3DMM 10 of a generic face is roughly shaped and fitted on top of the face in the image according to position, orientation, and large -i.e. low frequency- facial features. ii. A series of refinement steps are run to adjust the 3DMM to middle frequency, and then to high frequency facial details. b. The HRN algorithm estimates the position of multiple facial landmarks 20 including, but not limited to: i. the anterior nasal spine ii. the canthus lateralis right iii. the canthus lateralis left iv. the nasion

3. The estimated surface of the 3DMM from step 2a is converted into a three- dimensional point cloud 15 (see Fig. 5)

4. Using automatic anatomical mapping software (e.g. Brainlab’s Universal Atlas), the four anatomical landmarks mentioned in 2b are found in the patient image.

5. Using the anatomical landmarks 20 from 2b, the three-dimensional point cloud from step 3 is scaled to match the landmarks from the patient image 17. Then, a rough alignment 18 is conducted to rotate and translate 19 the three- dimensional patient image 17 to the scaled three-dimensional point cloud 15 from step 3 (illustrated in Fig. 6).

6. With the known positions of the facial landmarks in the three-dimensional point cloud, a bounding box is defined and the points outside this box are eliminated from both the pre-aligned point cloud.

7. A scale-adaptive fine alignment is done with a surface matching algorithm together with a scaling optimizer, which finds correspondences between the cropped point clouds of both the patient image and the three-dimensional point cloud from step 6.

8. If a match is found, the pose of the patient image is refined to match that of the three-dimensional point cloud from step 3 and the user is notified of the result of the match.

Fig. 7 is a schematic illustration of the medical system 4 according to the fifth aspect. The system is in its entirety identified by reference sign 4 and comprises a computer 5, a (for example non-transitory) program storage medium 8, a (for example non- transitory) electronic data storage device 6 for storing at least the patient data and a medical device 7 (such as a radiation treatment apparatus). The components of the medical system 4 have the functionalities and properties explained above with regard to the fifth aspect of this disclosure.

The present invention is associated with the following technical effects and/or advantages:

• A three-dimensional face reconstruction can be obtained from inputting only a two-dimensional RGB image.

• The positioning of three-dimensional data requires little or no user intervention and no additional markers.

• This process needs to be run once - after the pose of the patient’s face has been determined and the three-dimensional data has been overlaid, the pose of the three-dimensional data is kept in place using head pose tracking by the mixed reality device.

Claims

1. A computer-implemented method of establishing a spatial transformation between a two-dimensional digital camera image and a three-dimensional digital model of an anatomical body part, the method comprising the following steps: a) camera image data is acquired (S11 ) which describes a two-dimensional digital camera image of the anatomical body part; b) patient image data is acquired (S12) which describes a three-dimensional digital patient image of the anatomical body part; c) body part position data is determined (S13) based on the camera image data, wherein the body part position data describes a position of the image representation of the anatomical body part in the camera image; d) point cloud data is determined (S14) based on the body part position data, wherein the point cloud data describes a three-dimensional point cloud representation of the image representation of the anatomical body part and includes landmark point data describing the positions of points of the point cloud representing predetermined parts of the anatomical body part; e) landmark position data is determined (S15) based on the patient image data, wherein the landmark position data describes the positions of the image representations of the predetermined parts in the patient image; f) landmark matching data is determined (S16) based on the point cloud data and the landmark position data, wherein the landmark matching data describes a correspondence between the image representations of the predetermined parts in the patient image and each of the points of the point cloud representing the predetermined parts; g) landmark scaling data is determined (S17) based on the landmark matching data, wherein the landmark scaling data describes a relation between distances between the positions of each a pair of points of the point cloud representing predetermined parts and distances between the positions of each a pair of anatomically corresponding ones of the image representations of the predetermined parts in the patient image; h) scaled point cloud data is determined (S18) based on the landmark scaling data and the point cloud data, wherein the scaled point cloud data describes a scaling of the point cloud; i) transformation data is determined (S19) based on the scaled point cloud data and the landmark position data, wherein the transformation data describes a positional transformation between the positions of the points representing the predetermined parts of the anatomical body part in the scaling of the point cloud and the positions of the image representations of the predetermined parts of the anatomical body part in the patient image.

2. The method according to the preceding claim, wherein patient image transformation data is determined based on the patient image data and the transformation data, wherein the patient image transformation data describes a positional orientation of the patient image to the position of the anatomical body part, for example the result of an application of the positional transformation to the patient image, in a reference system of an imaging device used to generate the camera image data; and augmentation image data is determined based on the camera image data and the patient image transformation data, wherein the augmentation image data describes an image augmentation of the representation of the anatomical body part in the of the patient image and the camera image and/or at least part of a field of view of an imaging device used to generate the camera image.

3. The method according to the preceding claim, wherein camera tracking data is acquired which describes the position of an imaging device used to acquire the camera image data; and the augmentation image data is determined based on the camera tracking data and the transformation data.

4. The method according to any one of the two immediately preceding claims, wherein the image augmentation is displayed on a display device, wherein the display device and an imaging device used to generate the camera image data are integrated into the same device, for example a mixed reality device.

5. The method according to any one of the preceding claims, wherein scaling factor data is determined based on the landmark scaling data, wherein the landmark scaling data describes a plurality of relations between the distances between the positions of each a pair of points of the point cloud representing predetermined parts and distances between the positions of each a pair of anatomically corresponding ones of the image representations of the predetermined parts in the patient image, and the scaling factor data describes a mean scaling factor determined from the relations described by the landmark scaling data, wherein the scaled point cloud data is determined based on the scaling factor data.

6. The method according to any one of the preceding claims, wherein the body part position data is determined by inputting the camera image data into an algorithm configured to determine the position, in the camera image, of the image representation of the anatomical body part.

7. The method according to any one of the preceding claims, wherein the point cloud data is determined by inputting the camera image data into an algorithm configured to transform the image representation of the anatomical body part in the camera image into the point cloud and to determine the positions of the points of the point cloud representing the predetermined parts of the anatomical body part.

8. The method according to the preceding claim, wherein atlas data is acquired which describes a three-dimensional digital model of the anatomical body part and an identification of parts of the model corresponding to the predetermined parts of the anatomical body part, wherein the landmark position data is determined based on the patient image data and the atlas data.

9. The method according to any one of the preceding claims, wherein the two- dimensional camera image has been taken with an imaging device, for example a two-dimensional RGB camera, included in a mixed reality device.

10. The method according to any one of the preceding claims as far as dependent on claim 2, wherein the augmentation image data is used to determine the position of a situs on the anatomical body part for performing a medical procedure.

11 . The method according to any one of the preceding claims, comprising determining, based on the augmentation image data, control data for controlling a medical device, for example for controlling a medical device to attain a predetermined position relative to the anatomical body part or conduct a medical procedure on the anatomical body part.

12. The method according to any of the preceding claims, wherein the anatomical body part is or comprises the face, the spine or the pelvis.

13. The method according to any one of the preceding claims, wherein the predetermined parts are anatomical landmarks.

14. The method according to any one of the preceding claims, wherein the patient image data has been generated by applying a tomographic imaging modality such as computed x-ray tomography, magnetic resonance tomography or ultrasound tomography to the anatomical body part.

15. A computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method according to any one of the preceding claims; and/or a computer-readable storage medium on which the program is stored; and/or a computer comprising at least one processor and/or the program storage medium, wherein the program is executed by the processor; and/or a data carrier signal carrying the program; and/or a data stream comprising the program.

16. A medical system (4), comprising: a) the at least one computer (5) according to the preceding claim, wherein the program comprises instructions which, when the program is executed by a computer, cause the computer to carry out the method according to any one of the preceding claims as far as dependent on claim 3; b) at least one electronic data storage device (6) storing at least the patient image data; and c) a mixed reality device (7) for acquiring the camera image data and displaying the augmentation image data, wherein the at least one computer is operably coupled to the at least one electronic data storage device for acquiring, from the at least one data storage device, at least the patient image data, and wherein the at least one computer is operably coupled to or part of the mixed reality device for displaying the image augmentation on a display of the mixed reality device on the basis of the augmentation image data and the camera tracking data.