Computer-implemented method for surgical navigation

Abstract

The present invention relates to a computer-implemented method for surgical navigation comprising the following steps: obtaining image data of a patient, the image data having an image coordinate system; obtaining a position of a virtual coordinate system in the image, the virtual coordinate system having a spatial position and a spatial orientation in the image coordinate system; obtaining a spatial position of a patient sensor and a spatial orientation of the patient sensor in a tracker coordinate system, wherein the spatial position and orientation of the sensor are fixed in relation to the patient; obtaining at least a spatial position of a tracked instrument in the tracker coordinate system; and calculating a spatial position of the tracked instrument in the image coordinate system using the spatial position and spatial orientation of the virtual coordinate system and patient sensor.

Description

[0001] Computer-implemented method for surgical navigation

[0002] The present invention relates to a computer-implemented method for surgical navigation. Further, the present invention also relates to a method for surgical navigation using the computer-implemented method and to a surgical navigation system for carrying out the computer-implemented method.

[0003] To support surgeons during surgical procedures, e.g. in neurosurgery, ENT, oral and maxillofacial surgery, orthopedics, spine, interventional radiology, biopsy, etc., image-guided surgery (IGS) uses 3D surgical navigation in many cases. This uses preoperative radiological patient data sets (CT, MRI, PET, SPECT CBCT, etc.) and magnetic (EMTS) or optical (OTS) 3D measurement methods (tracker / navigation system technologies), which can measure and track corresponding sensors on a patient and on an instrument in three dimensions in real time. These measurements are usually displayed in the image space as crosshairs at the calculated position. This gives the surgeon a spatial orientation during the procedure, and he knows exactly where his tracked instruments are located in the patient's anatomy, even if the surgical site is not visible with the naked eye. In order to calculate and display the current position of a tracked pointing / surgical instruments at the patient's surgical site, the patient is correlated with his preoperative image data using the tracked instruments. This process is called patient-to-image registration or patient registration (see for example Horn B, Closed-Form Solution of Absolute Orientation Using Unit Quaternions, Journal of the Optical Society A (4) 4:629-642, doi: 10.1364 / JOSAA.4.000629, (1987)) and is usually performed in the operating room shortly before surgery. This procedure generally involves the use of radiopaque reference markers (registration points / markers / fiducials, at least four of which are required) that are attached or screwed to the body near the surgical site prior to imaging. The patient is then additionally scanned (CT, MRI, fluoroscopy, CBCT, etc.) while these are worn on / in the body. The 3D positions of registration markers are detected in patient data sets in the image space, localized, and their corresponding points on the patient's anatomy are targeted / touched intraoperatively with the tracked instrument.

[0004] Different registration methods (marker- and surface-based patient registration) are available, which have played an important role in IGS since the 1980s and form the core functionality of a surgical navigation system. Patient registration methods are clinically applicable with all clinically approved tracker technologies e.g. optical, magnetic, hybrid or others, or combinations of tracking technology.

[0005] Examples of registration methods are shown in the following documents:

[0006] • Fitzpatrick JM, Milan S, Handbook of Medical Imaging, Volume 2. Medical Image Processing and Analysis, Chapter 8: Image Registration PM80, ISBN: 9780819477606, (2009)

[0007] • P. J. Besl and N. D. Mckay, A Method for Registration of 3-D Shapes A Method for Registration of 3-D Shapes, IEEE Trans. Pattern Analysis and Machine Intelligence 1992 Vol. 14 Issue 2 Pages 239-256, http: / / dx.doi.org / 10.1109 / 34.121791.

[0008] • Horn B, Closed-Form Solution of Absolute Orientation Using Unit Quaternions, Journal of the Optical Society A (4) 4:629-642, doi: 10.1364 / JOSAA.4.000629, (1987)

[0009] A variety of surgical navigation and patient registration methods are known in the state of the art. A first type of patient registration methods are marker-based patient registration methods.

[0010] Marker-based registrations use either anatomical landmarks, dental splints, adhesive skin markers or invasive screw markers. With the anatomical landmark method (also known as manual registration), suitable soft tissue and bone points are first marked as registration points in the image data sets and then targeted intraoperatively on the patient using a tracked probe. Anatomical landmarks that can be quickly identified and precisely localized both in the image data and on the patient are considered particularly favorable. In addition, the points should be in as fixed a relationship as possible to the anatomical structures in the head area. From this point of view, the spina nasalis anterior, the tip of the nose, the medial and lateral canthi of the eye and the tips of the tragus are particularly suitable in the head region. The advantage of this method is the lack of need for additional (invasive or non- invasive) markers and additional imaging with the markers. This procedure is hardly associated with any additional stress for the patient and medical staff. However, the achievable accuracy is very limited, as the registration points can only be selected at specific locations on the head. Thus, the exact determination of the selected registration markers in the image data sets is sometimes problematic, which in some cases also means that the precise control of these virtually defined points on the patient's head is only possible with insufficient precision. The constantly varying patient anatomy presents the surgeon with new challenges with each registration procedure, so that certain routines and exercises are necessary to achieve optimal navigation accuracy. Changes in the patient's anatomy after imaging, such as skin swelling or shifts in the facial area, as well as the fact that the registration process may take longer than average, also have a limiting effect on this type of registration. The average accuracy of this procedure reported in the literature varies greatly from 0.9 mm to over 6 mm.

[0011] In the non-invasive dental splint method, registration markers are placed in a fixed and reproducible relationship to the patient's head using a customized dental splint. To ensure this, a variety of different models have been designed. One example is a dental splint that can be fixed to the patient's upper jaw using a vacuum. Other systems, such as a "Locking acrylic dental stent" (LADS), consist of several subunits for anterior and posterior teeth, which can be firmly connected to each other and thus ensure good stability. This contrasts with models formed from a single cast, such as the maxillary deep-drawing splint and the maxillary and oral vestibule silicone impression splint. The arrangement of the registration markers shows that a constellation of four alternately arranged markers has the highest navigational accuracy in the midface. The dental splint must be worn by the patient during imaging, can be removed afterwards and only needs to be reinserted immediately preoperatively. Provided that the anatomy of the dentition does not change, a dental splint that has been fitted once can be used several times if necessary. Studies have shown that dental splint registration with an accuracy of 1 to 2 mm is well suited for operations in the oral cavity, face, and orbit. However, with increasing distance from the registration markers, a significant reduction in precision with deviations of 3 to 5 mm can be observed in the cranial region. This problem was addressed with a frame construction attached to the dental splint, which can be fitted with additional registration markers placed close to the surgical site. This can enable more precise navigation in more distant areas such as the temporal region. A decisive factor for the resulting precision is the optimal positioning of the dental splint on the patient's teeth both during the imaging procedure and during intraoperative registration. Even minimal deviations from the correct placement can have a major impact on the navigation accuracy. The patient's dental status must also be taken into account when using this method. For example, age-related or traumatic changes and poor tooth stability or even the absence of teeth can have a restrictive effect on the use of a dental splint or make it completely impossible.

[0012] With the skin adhesive marker method, special adhesive markers are applied to the patient's skin before the image is taken, which then serve as registration markers. These must be easy to identify both in the image data sets as well as on the patient and have a structure, usually located in the center, which can be clearly identified as a registration point. With this in mind, many different marker systems have been developed, including radiopaque spheres, plastic tubes filled with contrast medium and multi-part products consisting of a base component attached to the skin and aluminum spheres that are adapted to it. As with the other marker-based registration methods, the number and distribution of the adhesive markers plays a decisive role in the resulting navigation accuracy. This varies in the literature between 0.80 mm and 3.86 mm. Limitations of this procedure include the movability of the skin. Changes in skin turgor, wrinkles and the patient's position can lead to a displacement of the adhesive markers and make registration difficult or impossible. There is also a risk of the markers becoming detached between image acquisition and the surgical procedure, whereby attempts are sometimes made to enable reproducible positioning of the markers using visual markers.

[0013] In an invasive screw marker method, a varying number of titanium screws (at least 4) are distributed and anchored in the patient's skull bone under local anesthesia, which then serve as registration markers and remain in situ until after the surgical procedure. These registration points can be clearly identified both in the image data set and on the patient and, due to their unchanged position in the time interval between imaging and surgery, allow a very precise correlation and therefore very precise accuracy between virtual and real patient anatomy. Even changes in the patient's physiognomy due to, for example, position-related displacement of the skin surface, swelling (e.g. edema, hematomas, etc.) or injuries in the facial area have no influence on the correct detection of the registration markers. The average accuracy of this procedure is stated in the literature to be between 0.67 and 1.10 mm. As with any invasive procedure, the attachment of multiple screw markers carries the risk of wound infection, injury to anatomical structures such as nerves and vessels and potential secondary bleeding. Possible skin irritation and residual scarring are potential disadvantages of this method.

[0014] All types of registration fiducials / markers are known in the field are extrinsic markers.

[0015] Registration methods based on surface recognition offer an alternative to marker-based methods. The basis for this concept of surface matching is a process also known as segmentation. Using image data sets, 3D reconstructions, in this case of the patient's head, especially the face, can be generated by the computer. This virtual 3D model is then correlated with the recorded facial contours of the real patient, usually using the "Iterative Closest Point" (ICP) algorithm. All registration methods based on surface recognition have certain limitations. Changes in the surface structure of the patient's face between data acquisition and surgical intervention (e.g. edema or changes in facial expression) have a decisive influence on the resulting navigation accuracy. For this reason, it is hardly possible to repeat intraoperative registration procedures if the patient's anatomy deviates too much from the previously acquired image data as a result of the procedure, which consistently leads to errors that are twice as large as when using point pair alignment with adhesive markers.

[0016] In a laser-based method, surface data is captured using a visible laser beam. In this semiautomatic procedure, the facial contours of the patient are scanned by the surgeon as if using a pointer instrument. The aim is to capture the most prominent, individual structures of the patient's anatomy, which - hairless and only covered by a thin layer of soft tissue -reflect the profile well. Examples of such regions are the bridge of the nose, the nostris, the orbitae, and the forehead area. The reflected laser beams are detected by a camera system and a virtual, three-dimensional profile of the patient's anatomy can be calculated with the aid of a computer using the surface points recorded. The average accuracy of this procedure is stated in the literature to be between 1.0 - 2.8 mm, although deviations of up to 9 mm have also been reported.

[0017] With pointing instrument-based surface detection methods, the most prominent facial contours of the patient are traced with a tracked probe and a varying number of surface points are registered. The same criteria apply to the selection of structures to be included as for laser-based surface recognition. The accuracy of this method is given in the literature as between 1.81 and 4.90 mm. One difficulty in performing this procedure is the handling of the pointing instrument. A surgeon must not exert too much pressure on the soft tissue during the registration process, nor hover in the air without contact with it.

[0018] The LED mask method is an alternative to the pointer instrument-based surface detection method. A self-adhesive, flexible mask equipped with LEDs is attached to the patient's face in a standardized manner. The LEDs are supplied with power via a battery-powered communication unit connected to the mask. Flexible copper plates serve as conductor layers. The LEDs can be distributed over the facial area and emit infrared rays. The LEDs are captured by a high-resolution camera system and transmitted to a computer workstation. This data can then be used to create a three-dimensional reconstruction of the patient's anatomy. The camera system, registration masks and communication unit not only register the patient, but also record changes in head position. The average overall accuracy of this system is stated in the literature to be between 2.22 and 3.65 mm. As with all optical procedures, care must be taken when using the LED mask to ensure that the so-called "line-of-sight" between the mask and 3D camera is permanently guaranteed. Factors such as sweat, blood, secretions, rinsing, etc., which can cause the mask to become detached, can also have a negative effect on consistently high perioperative navigation accuracy. All of the patient registration methods that use medical imaging of extrinsic registration markers to realize surgical navigation need extra medical imaging, leading to additional radiation exposure, extra pre-operative time and costs. Moreover it is known that various effects like localization errors of markers in images and on patients, tracker error, registration errors, etc. affect the application accuracy of surgical navigation.

[0019] The present invention aims to address one or more of the above-mentioned disadvantages. In particular, the present invention seeks to provide a simple registration and tracking method for following the position of a tracked probe or any other surgical instrument.

[0020] A first aspect of the present invention relates to a computer-implemented method for surgical navigation. In a step of the method, image data of a patient is obtained. The image data has an image coordinate system. As mentioned above, the image data may be volumetric or non-volumetric. Example image data, e.g. volumetric image data is CT, MRI, PET, SPECT, CBCT, or 3D ultrasound image data. Non-volumetric image data may be ultrasound or fluoroscopy image data The coordinate system may be included in the image data during recording or, in some embodiments, be added later on. In particular, the image data may be 3D. For example the image data may comprise a plurality of images that are adjacent slices. The image data may be received by a software or hardware interface. In one example, the image data is stored in a central server and then accessed by a computer. In other examples, the image data may be stored on a volatile or nonvolatile memory of a computer or computing instance implementing the method described herein.

[0021] In a further step, a position of a virtual coordinate system in the image is obtained. The virtual coordinate system may have a spatial position and may have a spatial orientation in the image coordinate system. The virtual coordinate system may be manually or automatically set as will be described in further detail below. The virtual coordinate system may be placed anywhere in the image data. Preferably, the virtual coordinate system is placed close to or in the origin of the image coordinate system, or close to one of the primary anatomical target(s) of interest. Thus the transformation of coordinates from the virtual tracker coordinate system to the patient image coordinate system is exactly known. The virtual coordinate system may be used in calculating the current position of a tracked instrument during an operation, in particular a computer-guided operation. In alternative or additional embodiments, the position may also be tracked prior to and / or after an operation.

[0022] The position of a virtual coordinate system in the image may be received from another device or may be defined in the method.

[0023] Spatial position as used herein may refer to the position of objects in space. Spatial orientation may refer to an orientation of objects in space.

[0024] In a further step, a spatial position of a patient sensor and a spatial orientation of a patient sensor in a tracker coordinate system is obtained. The spatial position and the spatial orientation of the patient sensor are fixed in relation to a patient. The patient sensor allows following the patient position and orientation in space in case the patient moves or is moved during operation. In one example, the patient sensor may be the sensor disclosed in PCT / EP2023 / 077822 although other patient sensors may also be suitable. The spatial position and spatial orientation of the patient sensor may be obtained with a tracking system having a tracker coordinate system. A tracking system may be a known surgical 3D tracking system. The patient sensor may be a physical sensor or reference frame.

[0025] In a further step, at least a spatial position of a tracked instrument is obtained in the tracker coordinate system. The tracked instrument may be a probe, a surgical tool, or any other instrument that can be tracked with the tracking system. The position of the tracked instrument may be determined using sensors (optical (OTS) and electromagnetic (EMTS) tracking sensors) from any tracking system or from any other suitable 3D tracking system. As is known in the art, the tracking system and probe may need to be calibrated so that coordinates originating from the sensor are correctly mapped to the tool tip. Further, the position of the sensor on the tool should not change so the calibration remains accurate.

[0026] In a further step, a spatial position of the tracked instrument in the image coordinate system is calculated. The tracked instrument spatial position in the image coordinate system may be calculated using the spatial position and spatial orientation of the virtual coordinate system, the image coordinate system and the tracker coordinate system. The calculation may involve transformation matrices that transform the position of the tracked instrument from the tracker coordinate system to the image coordinate system.

[0027] The above steps may be executed in the given order or in a different order. For example, the steps a. to d. as outlined below may be executed in any order.

[0028] The present invention allows for an easy and direct tracking of the tracked instrument in the image coordinate system, e.g., without needing any patient-to-image registration process as described above. In particular, using a virtual coordinate system allows for a transform to the image coordinate system without loss in precision. Moreover, there is no need for an explicit patient-to-image-registration.

[0029] In a preferred embodiment, one, two, three or more virtual coordinate systems are placed. If more than one virtual coordinate system is placed, these virtual coordinate systems need not have six degrees-of-freedom and the spatial orientation of the resulting virtual coordinate system may be defined implicitly.

[0030] The step of obtaining a spatial position of the patient sensor and a spatial orientation of the patient sensor may comprise obtaining the spatial position and orientation of a single patient sensor. In principle, more than one patient sensor could be used. However, it may be sufficient to calculate the coordinate transformation of the position of the tracked instrument from the tracker coordinate system to the image coordinate system using only a single patient sensor.

[0031] In a preferred embodiment, the patient sensor is a 6 degree of freedom, 6-DOF, sensor. This may mean that the sensor may record movement, i.e. movement of the patient, in all spatial directions (e.g. in X, Y, Z directions, wherein the X, Y, Z directions may be orthogonal to each other) and rotation (rotation around X, Y, Z directions, wherein the X, Y, Z directions may be orthogonal to each other) around all spatial directions. The tracker coordinate system and / or the image coordinate system may preferably employ a Cartesian or any other suitable coordinate system.

[0032] Preferably, the virtual coordinate system has six degrees of freedom, 6-DOF. Thereby, only a single virtual coordinate system may be used. But multiple instances of virtual coordinate systems may be beneficial.

[0033] Preferably, the method comprises the step of obtaining a transformation of the position of a tracked probe in the tracker coordinate system into the image coordinate system via a patient sensor in the tracker coordinate system and calculating a spatial position of the patient sensor in the image coordinate system using this coordinate transformation and with a virtual coordinate system. Thereby, a position can be transformed from the tracker coordinate system to the image coordinate system.

[0034] The coordinate transformation between the position of the virtual coordinate system in the image coordinate system and the physical marker (patient tracker) in the tracker coordinate system may be particularly simple, if the virtual coordinate system is chosen to be in the same position and / or has the same orientation as the physical sensor.

[0035] In an alternative embodiment the virtual coordinate system is chosen such that it is in the origin of the tracker coordinate system. The virtual coordinate system may be placed in the origin of the image coordinate system. Thereby, the coordinate transformation and the transformation is simplified.

[0036] More generally, the position of the virtual coordinate system may be chosen anywhere in the image coordinate system.

[0037] In a preferred embodiment, the method may comprise the step of obtaining a coordinate transformation between a position of the tracked instrument in the tracker coordinate system and a position of the patient sensor in the tracker coordinate system. Calculating at least a spatial position of the tracked instrument, preferably a spatial position and a spatial orientation, may use this coordinate transformation. Preferably, the method comprises the step of obtaining a coordinate transformation between tracked instrument in the virtual coordinate system in the image coordinate system and the tracked instrument in the tracker coordinate system. Calculating a spatial position of the tracked instrument in the image coordinate system may use this coordinate transformation.

[0038] In a preferred embodiment, the coordinate transformation may include a single coordinate transformation matrix in particular a suitable transformation matrix for transforming the position of trackers in the tracker coordinate system directly into a position of the tracker in the image coordinate system. Thereby, the above-mentioned coordinate transformations may be obtained in one integrated step.

[0039] Preferably, the position of the virtual coordinate system may be indicated by a user. For example, a user may indicate with a cursor, e.g. using a mouse or touchscreen or any other input methods / devices, where the virtual coordinate system should be positioned. In certain embodiments, a user may also indicate an orientation of the virtual coordinate system.

[0040] In other embodiments, the method may include a step of generating a position and / or orientation of the virtual coordinate system, e.g. automatically. In certain embodiments, the generated position may be displayed to a user who may or may not confirm the suggested position and / or orientation of the virtual coordinate system. Preferably, the position and / or orientation of the virtual coordinate system is automatically generated by computing means.

[0041] In a preferred embodiment, the image data comprises medical image data, e.g radiological image data. In particular the image data may be obtained with at least one of a CT, a MRT, a PET, a SPECT, a CBCT, a 3D US, a PET-CT or similar volumetric or non-volumetric imagery. The image data may also comprises a number of non-volumetric image data sets suitable for generating volumetric data. In one example non-volumetric image data sets are ultrasound image data sets.

[0042] Preferably, the patient sensor is a magnetic or optical sensor or any other kind of trackable sensor. In a preferred embodiment, the method does not require registration fiducials or markers to be attached to the patient. Basically, the 6-DOF sensor acts as a reference since its position is uniquely set in the tracker coordinate system, i.e. it has some of the functions of a fiducial on a patient. However, the system does not have any separate fiducials with a separate registration as the prior art. In some embodiments, the position of the patient sensor may be known in relation to the patient. Thus, based on the position of the patient sensor, a coordinate transformation between the patient sensor and the image coordinate system sensor may be obtained straightforwardly.

[0043] The patient sensor may be directly attached to the patient. The patient sensor may have a fixed spatial and orientation relationship to a target region. For example, in neurosurgery, the target region may be within the cranium. In this case, the patient sensor may be fixedly attached to one of the skull bones.

[0044] In alternative embodiments, the patient sensor may not be directly attached to the patient but may at least have a fixed spatial and / or orientation relationship to the patient, or at least to a target region of a patient. For example, in neurosurgery, the patient's head may be fixed to an operation table. In this case, it may also be sufficient if the patient sensor has a fixed spatial (and orientational) relationship to the operation table.

[0045] Additionally to the above, the method may comprise one or more of the following steps: displaying the virtual coordinate system in the image data, displaying the tracked instrument in the image data, and / or displaying the patient sensor in the image data, and / or displaying a surgical target in the image data. This may facilitate an operation. A surgeon may track the current position of the tracked instrument in the image data which allows for more precise navigation, especially when the surgeon is operating in the region that is not directly visible, e.g. in case of minimally invasive surgeries.

[0046] The image data may be two-dimensional, although, as mentioned above, the image data is preferably three-dimensional. Additionally to the position of the tracked instrument, an orientation of the tracked instrument may also be calculated in the image coordinate system. In this case, an orientation of the tracked instrument in the tracker coordinate system may be obtained (e.g. recorded with the surgical navigation system) and then transformed into the image coordinate system similarly to the position of the tracked instruments. Thereby, not only a position, e.g. of a tip of the tracked instrument may be obtained (and e.g. shown), but also an orientation. In some embodiments this may be used to show a schematic version of the tracked instrument in the image data. This may allow a surgeon to see the full instrument rather than only one point of the instrument in the image data, e.g. during operation.

[0047] A further aspect of the present invention relates to a method for surgical navigation. The method comprises the step of fixating a sensor in the field of view of a 3D tracking device of a surgical navigation system. In particular, the sensor may be fixed on or near a patient. The sensor may be fixed on or near the patient in a nonsurgical manner. A spatial position and orientation of a patient sensor in the tracker coordinate system may be detected. Using at least this data, the computer-implemented method as described above may be executed.

[0048] Preferably, the sensor is fixed on a suitable bodily part upon which surgery is executed, in particular a cranium, a neurocranium, a maxilla, a spine, a hip, a chest, or an abdomen.

[0049] Preferably, the tracked instrument is a probe.

[0050] A further aspect of the present invention relates to a data processing system comprising means for carrying out the method as described above.

[0051] A further aspect of the present invention rates to a computer program comprising instructions which, when the program is executed by a computer or computing instance, cause the computer to carry out the method as described above.

[0052] A further aspect of the present invention relates to a computer readable medium comprising instructions which, when executed by a computer, cause the computer to carry out the method as described above. A further aspect of the present invention relates to a surgical navigation system comprising means for carrying out the method as described above.

[0053] The present invention will be described with respect to an exemplary embodiment, that is not construed to be limiting the scope of the claims in the following.

[0054] Figure 1 shows image data, in which a virtual coordinate system, vDRF, in the image coordinate system is placed.

[0055] Figure 2 shows, without loss of generality, as a variety of representations of matrices are known in the field, an example of a matrix for transforming the position of the virtual coordinate system (in the image coordinate system) to a tracker coordinate system and vice versa (via the inverse).

[0056] Figure 3 shows an embodiment of the present invention using two exemplary positionings of physical sensors as patient trackers.

[0057] Figure 4 shows, without loss of generality, as a variety of representations of matrices are known in the field, an example of a transformation matrix between patient sensor and tracker coordinate system.

[0058] Figure 5 shows a tracked instrument according to the invention where a position, and ideally orientation, sensor is placed juxtaposed to the of the instrument's tip.

[0059] Figure 6 shows, without loss of generality, as a variety of representations of matrices can are known in the field, an example of a transformation matrix between a tooltip via a tool sensor on a tracked instrument (it) according to figure 5 and the tracker coordinate system.

[0060] Figure 7 shows a chain of coordinate transformations according to figures 2, 4 and 6.

[0061] Figures 8 to 11 illustrate the chain of transformations according to figure 7.

[0062] Figure 1 shows images obtained by a CT. Figure 1 shows three cross-sections of a cranium of a patient. On the left, figure 1 shows an axial view (z-axis perpendicular to the image slice shown) of a volumetric medical image and the according image coordinate system. In the center, a sagittal view (y-axis perpendicular to the image slice shown) of a volumetric medical image and the according image coordinate system is shown. On the right, a coronal view (x-axis perpendicular to the slice shown) of a volumetric medical image and the according image coordinate system is shown. In all of these mutually orthogonal crosssections, a virtual coordinate system, vDRF, 10 is placed. The vDRF is oriented within the image coordinate system such that a first value may define an orientation, i.e. a rotation around a first axis, e.g. the x-axis. A second value may define an orientation, i.e. a rotation around a second axis, e.g. the y-axis. A third value may define an orientation, i.e. a rotation around a third axis, e.g. the z-axis.

[0063] The virtual coordinate system 10 may be automatically placed, e.g. using a computer executing the method described herein, or manually. If the virtual coordinate system 10 is manually placed, a user may indicate a position and / or orientation using a touchscreen and / or a cursor with an input device (e.g. a mouse). In some embodiments, a computer may calculate a position and / or orientation of the virtual coordinate system as will be described in further detail below and ask a user to confirm the position and / or orientation or input an alternative position and / or orientation manually (e.g. in a prompt).

[0064] According to embodiments, the present invention may include a surgical navigation system. Surgical navigation systems are known as such. Example surgical navigation system are Brainlab Kolibri or Curve, Medtronic Stealth Station, or Stryker Q Guidance or ENT Navigation System.

[0065] The surgical navigation system may comprise a patient sensor. Figure 3 shows a first patient sensor 20 schematically. Patient sensors are known in the art and may be magnetic or optical or any other sensing systems. The surgical navigation system may detect a current position (and optionally additionally a current orientation) of the first patient sensor 20. In some embodiments, the surgical navigation system may detect a current position and current orientation of the patient sensor (6 degree of freedoms sensor). In another embodiment the surgical navigation system may detect the position of several patient sensors and calculate an orientation based on the plurality of patient sensors. The patient sensor has a fixed spatial and orientation or relationship to a patient. In the example shown in figure 3, the first patient sensor 20 is fixedly attached to a skull of a patient and is a 6 degree of freedom sensor. The coordinates of the first patient sensor 20 are uniquely defined in the tracker coordinate system, i.e. the surgical navigation system can detect the position and orientation of the first patient sensor 20. The position and orientation of the first patient sensor 20 may be obtained in a coordinate system defined by the surgical navigation system, which may be referred to herein as a tracker coordinate system.

[0066] Alternatively (or additionally) to the first patient sensor, the surgical navigation system may comprise a second patient sensor 30. The second patient sensor 30 is not in direct contact with the patient, but connected to the patient, for example using a mechanical arm. Similar to the second patient sensor 30, the surgical navigation system may obtain, in the tracker coordinate system, a position and an orientation of the second patient sensor 30.

[0067] An example position may be expressed in XYZ coordinates and an example orientation may be defined in quaternions, e.g. qx, qy, qz, qw. Quaternions are particularly useful for describing spatial rotations, i.e. spatial orientations. The use of quaternions in the following serves exemplary purposes as a variety of other representations of transformations are available to those learned in the field. The XYZ coordinates may be expressed as follows:

[0068] • x: sensor position in x-axis

[0069] • y: sensor position in y-axis

[0070] • z: sensor position in z-axis

[0071] The quaternions may be expressed as follows:

[0072] • qx: unit vector, multiplied by si n (a ngle / 2) to determine the orientation / rotation of the sensor in x-axis

[0073] • qy: unit vector, multiplied by sin(angle / 2) to determine the orientation / rotation of the sensor in y-axis • qz: unit vector, multiplied by sin(angle / 2) to determine the orientation / rotation of the sensor in z-axis

[0074] • qw: unit vector for fourth dimension, multiplied by cos(angle / 2) to determine the amount of rotation towards an axis

[0075] Based on the quaternions, a rotation matrix may be devised. An example rotation matrix is shown as part of the matrix of figure 4. The rotation matrix is a 3 x 3 matrix that expresses how the patient sensor 20 or the patient sensor 30 is rotated with respect to the tracker coordinate system. Further, the matrix of figure 4 also includes a 3 x 1 translation vector that expresses a position of the first patient sensor 20 or the second patient sensor 30 as seen from the origin of the tracker coordinate system. Lastly, figure 4 includes a 1 x 4 dimensional vector that may be used to scale the first physical sense of 20 or the second patient sensor 30. However, since the position of the first patient sensor 20 or the second patient sensor 30 is directly obtained in the tracker coordinate system, scaling is usually not necessary. As a result, as expressed in figure 4 there is usually no scaling.

[0076] Similarly, the surgical navigation system may determine a position and orientation of a tracked instrument such as probe 40 (see figure 5) using a tracker 50 (see figure 8). The surgical navigation system 1 may detect a current position and orientation of the probe 40 using XYZ coordinates, and quaternions similarly to the sensors as described above with the tracker 50. This positional information may be expressed based on the tracking coordinate system. The information measured by the probe 40 may be used to calculate a position of the probe 40 as seen from the origin of the tracking coordinate system using the matrix shown in figure 6, e.g. relative to the tracker 50. The position of the probe 40 may be the position of a tip 41 of the probe and an orientation of the probe as a whole.

[0077] In another example, the position and orientation of the virtual coordinate system 10 may correspond to the origin of the axis of the tracker coordinate system and an orientation of the axis of the tracker coordinate system. The virtual coordinate system 10 may form a (new) origin of the image coordinate system) and may define a new orientation of the image coordinate system, e.g. the coordinate axis may be rotated. This may also be a particularly simple case as tracker and image coordinate systems will become substantially identical and tracker image coordinates can be used interchangeably and may only need scaling.

[0078] Based on the above, a coordinate transform that calculates a position of the probe 40 in the image coordinate system may be expressed as shown by the equation in figure 7, in which:

[0079] • T: is a rigid body transformation, e.g. one of the matrices shown in figures 2, 4, and 6.

[0080] • t: refers to the tracker coordinate system

[0081] • it: Tracked instrument position in tracker coordinate system

[0082] • tDRF: 6-DOF patient sensor position in tracker coordinate system

[0083] • VDRF: 6-DOF virtual coordinate system position in image coordinate system

[0084] • ic: Tracked instrument position in image coordinate system

[0085] Thus, the following matrices are expressed in fig.7:

[0086] •lcTitA transformation for the position and orientation of the probe 40, (it) as measured by the tracker 50 in the tracker coordinate system to the position and orientation of the probe 40 in the image coordinate system (ic).

[0087] •lc^vDRF- transformation from the virtual coordinate system 10 (VDRF) to the image coordinate system (ic).

[0088] •VDRFTtDRF:transformation for the position and orientation of the patient sensor in the tracker coordinate system (tDRF) to the position and orientation of the patient sensor to the virtual coordinate system (VDRF).

[0089] •tDRFTtA transformation for the position and orientation in the tracker coordinate system(t) to position and orientation in the patient sensor (tDRF) (20 and / or 30) coordinate system.

[0090] •fTit: A transformation for the position and orientation of the probe 40 (it) as measured by the tracker in the tracker coordinate system 50, (t).

[0091] The mathematical expression shown in fig. 7 is illustrated in figures 8 to 11 graphically.

[0092] Figure 8 shows the transformation of the orientation and position of the probe 40 to the origin of the tracker coordinate system as represented by surgical navigation system 1 as illustrated by the arrow (i.e.

[0093]

[0094] fTit). Figure 9 then adds the coordinate transformation according to the inverse oftDRFTt. Figure 10 adds to figure 9 the transformationVDRFTtDRFand figure 11 adds1CTVDRF. AS shown in figure 11, the system may display a position of the probe tip 41 in the image data, e.g. using crosshairs 2 and a line symbolizing the axis of a probe without excluding other visualization paradigms of positions and probes in medical data sets.

[0095] In the following, particular simplified numerical examples of a transformation as described above with reference to figures 7 to 11 will be shown. The probeToTrackerTransformation isfTit, the patientDRFToTrackerTransformation istDRFTt-1: , and opatientDRFToVirtualDRFTransformation is

[0096] VDRFTtDRF, virtualDRFTolmageCoordinateTransformation corresponds to the matrix1CTVDRF

[0097] In a first example, a probe is translated along the x-axis by 10 mm while the patient sensor (also patient DRF) is in a constant position.

[0098] 1 0 0 10'

[0099] 0 1 0 0 probeToTrackerTransformation:

[0100] 0 0 1 0

[0101] .0 0 0 1 .

[0102] 1 0 0 0'

[0103] 0 1 0 0 patientDRFToTrackerTransformation:

[0104] 0 0 1 0

[0105] .0 0 0 1.

[0106] 1 0 0 0'

[0107] 0 1 0 0 patientDRFToVirtualDRFTransformation:

[0108] 0 0 1 0

[0109] .0 0 0 1.

[0110] 1 0 0 10'

[0111] 0 1 0 0 virtualDRFTolmageCoordinateTransformation:

[0112] 0 0 1 0

[0113] .0 0 0 1 .

[0114] The probe is moved 10 mm in x-axis direction in the image coordinate system. In a second example, a probe is rotated by 90° around z-axis while the patient sensor (also patient DRF) is in a constant position.

[0115] 0 -1 0 O'

[0116] 1 0 0 0 probeToTrackerTransformation:

[0117] 0 0 1 0

[0118] .0 0 0 1.

[0119] 1 0 0 0'

[0120] 0 1 0 0 patientDRFToTrackerTransformation:

[0121] 0 0 1 0

[0122] .0 0 0 1.

[0123] 1 0 0 0'

[0124] 0 1 0 0 patientDRFToVirtualDRFTransformation:

[0125] 0 0 1 0

[0126] .0 0 0 1.

[0127] 0 -1 0 O'

[0128] 1 0 0 0 virtualDRFTolmageCoordinateTransformation:

[0129] 0 0 1 0

[0130] .0 0 0 1.

[0131] The probe is rotated 90° around the z-axis in the image coordinate system.

[0132] In a third example, a probe is placed at 5 mm in z-axis while the patient sensor (also patient DRF) is translated in x-axis by 10 mm.

[0133] 1 0 0 0'

[0134] 0 1 0 0 probeToTrackerTransformation:

[0135] 0 0 1 5

[0136] .0 0 0 1.

[0137] 1 0 0 10'

[0138] 0 1 0 0 patientDRFToTrackerTransformation:

[0139] 0 0 1 0

[0140]

[0141] .0 0 0 1 . 1 0 0 0'

[0142] 0 1 0 0 patientDRFToVirtualDRFTransformation:

[0143] 0 0 1 0

[0144] .0 0 0 1.

[0145] 1 0 0 -10'

[0146] 0 1 0 0 virtualDRFTolmageCoordinateTransformation:

[0147] 0 0 1 5

[0148]

[0149] .0 0 0 1 .

[0150] The probe position remains constant in the image coordinate system while the patient sensor changes its position.

[0151] In a fourth example, a probe position is constant while the patient sensor (also patient DRF) is rotated by 180° around the x-axis.

[0152] 1 0 0 0'

[0153] 0 1 0 0 probeToTrackerTransformation:

[0154] 0 0 1 0

[0155] .0 0 0 1.

[0156] 1 0 0 O'

[0157] 0 -1 0 0 patientDRFToTrackerTransformation:

[0158] 0 0 -1 0

[0159] .0 0 0 1.

[0160] 1 0 0 0'

[0161] 0 1 0 0 patientDRFToVirtualDRFTransformation:

[0162] 0 0 1 0

[0163] .0 0 0 1.

[0164] 1 0 0 O'

[0165] 0 -1 0 0 virtualDRFTolmageCoordinateTransformation:

[0166] 0 0 -1 0

[0167] .0 0 0 1.

[0168] The probe position remains constant, but the orientation is changed 180° around the x-axis in the image coordinate system.

[0169] The present invention also relates to the following aspects: 1. Computer-implemented method for surgical navigation comprising the following steps:

[0170] a. Obtaining image data of a patient, the image data having an image coordinate system;

[0171] b. Obtaining a position of a virtual coordinate system (10) in the image, the virtual coordinate system (10) having a spatial position and a spatial orientation in the image coordinate system;

[0172] c. Obtaining a spatial position of a patient sensor (20, 30) and a spatial orientation of the patient sensor (20, 30) in a tracker coordinate system, wherein the spatial position and orientation of the sensor are fixed in relation to the patient;

[0173] d. Obtaining at least a spatial position of a tracked instrument in the tracker coordinate system; and

[0174] e. Calculating a spatial position of the tracked instrument in the image coordinate system using the spatial position and spatial orientation of the virtual coordinate system (10) and patient sensor (20, 30).

[0175] 2. Computer-implemented method according to aspect 1, wherein one, two, three or more virtual coordinate systems (10) are placed.

[0176] 3. Computer-implemented method according to aspect 1 or 2, wherein the spatial position and orientation of a single patient sensor to track the patient is obtained in step c.

[0177] 4. Computer-implemented method according to one of the preceding aspects, wherein the patient sensor (20, 30) is a preferably six-degree of freedom, 6-DOF, sensor or any suitable combination of sensors with less degrees-of-freedom.

[0178] 5. Computer-implemented method according to one of the preceding aspects, wherein the virtual coordinate system (10) has six degrees of freedom, 6-DOF, or any suitable combination of sensors with less degrees-of-freedom.

[0179] Computer-implemented method according to one of the preceding aspects, comprising the step of obtaining a coordinate transformation between a virtual coordinate system (10) in an image coordinate system and a patient sensor (20, 30) in a tracker coordinate system to transform spatial positions of a tracked instrument into the image coordinate system.

[0180] Computer-implemented method according to one of the preceding aspects, comprising the step of obtaining a coordinate transformation between a position of the tracked instrument in the tracker coordinate system and a position of the patient sensor (20, 30) in the tracker coordinate system and wherein calculating at least a spatial position of the tracked instrument, preferably a spatial position and a spatial orientation, in the image coordinate system uses this coordinate transformation.

[0181] Computer-implemented method according to one of the preceding aspects, comprising the step of obtaining a coordinate transformation between the virtual coordinate system (10) in the image coordinate system and the position of the tracked instrument in the image coordinate system and wherein calculating a spatial position of the tracked instrument in the image coordinate system uses this coordinate transformation.

[0182] Computer-implemented method according to one of the aspects 6 to 8, wherein the coordinate transformation includes a suitable coordinate transformation well known to those learned in the field.

[0183] Computer-implemented method according to one of the preceding aspects, wherein obtaining a position of the virtual coordinate system (10) may include at least one of: receiving a user input indicating a position and / or orientation of the virtual coordinate system (10), determining a position of the virtual coordinate system (10), and generating a position and / or orientation of the virtual coordinate system (10). Computer-implemented method according to one of the preceding aspects, wherein the image data comprises radiological image data.

[0184] Computer-implemented method according to one of the preceding aspects, wherein the image data is obtained with one or more of such as: a CT, a MRT, a PET, a SPECT, and a CBCT, 3D US, PET-CT or similar volumetric radiologic imagery.

[0185] Computer-implemented method according to one of the preceding aspects, wherein the patient sensor (20, 30) is a magnetic or optic sensor or any other suitable position and orientation sensing device.

[0186] Computer-implemented method according to one of the preceding aspects, wherein performing the method does not require attaching registration fiducials to the patient.

[0187] Computer-implemented method according to one of the preceding aspects comprising at least one of the following steps: displaying the virtual coordinate system (10) in the image data, displaying the tracked instrument in the image data, displaying the patient sensor (20, 30) in the image data, and displaying a surgical target in the image data.

[0188] Computer-implemented method according to one of the preceding aspects, wherein the image data is at least two-dimensional.

[0189] Computer-implemented method according to one of the preceding aspects, wherein an orientation of the tracked instrument in the image coordinate system using the spatial position and spatial orientation of the virtual coordinate system (10) and patient sensor (20, 30) is calculated.

[0190] Method for surgical navigation comprising the steps of fixating a sensor in a field of view of a surgical navigation system, in particular on or near a patient, detecting a spatial position of a tracked instrument in the tracker coordinate system, and executing the computer-implemented method according to one of the preceding aspects.

[0191] 19. Method for surgical navigation according to aspect 19, wherein the sensor is fixated on a bodily part, in particular, a cranium, a neurocranium, a maxilla, a spin, a hip, a chest, or an abdomen.

[0192] 20. Method for surgical navigation according to aspect 19, wherein the tracked instrument is a probe (40).

[0193] 21. A data processing system comprising means for carrying out the method of any one of aspects 1 to 18.

[0194] 22. A computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method of any one of aspects 1 to 18.

[0195] 23. A computer-readable medium comprising instructions which, when executed by a computer, cause the computer to carry out the method of any one of aspects 1 to 18.

[0196] 24. A surgical navigation system comprising means for carrying out the method of any one of aspects 1 to 18.

Claims

Claims1. Computer-implemented method for surgical navigation comprising the following steps:a. Obtaining image data of a patient, the image data having an image coordinate system;b. Obtaining a position of a virtual coordinate system (10) in the image, the virtual coordinate system (10) having a spatial position and a spatial orientation in the image coordinate system;c. Obtaining a spatial position of a patient sensor (20, 30) and a spatial orientation of the patient sensor (20, 30) in a tracker coordinate system, wherein the spatial position and orientation of the sensor are fixed in relation to the patient;d. Obtaining at least a spatial position of a tracked instrument (40) in the tracker coordinate system; ande. Calculating a spatial position of the tracked instrument in the image coordinate system using the spatial positions and spatial orientations of the virtual coordinate system (10) and patient sensor (20, 30), respectively.

2. Computer-implemented method according to claim 1, wherein the spatial position and orientation of a single patient sensor to track the patient is obtained in step c.

3. Computer-implemented method according to one of the preceding claims, wherein the patient sensor (20, 30) is a six-degree of freedom, 6-DOF, sensor or any suitable combination of sensors with less than six degrees-of-freedom.

4. Computer-implemented method according to one of the preceding claims, wherein the virtual coordinate system (10) has six degrees of freedom, 6-DOF.

5. Computer-implemented method according to one of the preceding claims, comprising the step of obtaining a coordinate transformation between the virtual coordinate system (10) in the image coordinate system and the patient sensor (20, 30) in the tracker coordinate system.

6. Computer-implemented method according to one of the preceding claims, comprising the step of obtaining a coordinate transformation between a position of the tracked instrument in the tracker coordinate system and a position of the patient sensor (20, 30) in the tracker coordinate system and wherein calculating at least a spatial position of the tracked instrument, preferably a spatial position and a spatial orientation, in the image coordinate system uses this coordinate transformation.

7. Computer-implemented method according to one of the preceding claims, comprising the step of obtaining a coordinate transformation between the virtual coordinate system (10) in the image coordinate system and the tracked instrument in the image coordinate system and wherein calculating a spatial position of the tracked instrument in the image coordinate system uses this coordinate transformation.

8. Computer-implemented method according to one of the preceding claims, wherein obtaining a position of the virtual coordinate system (10) may include at least one of: receiving a user input indicating a position and / or orientation of the virtual coordinate system (10), determining a position of the virtual coordinate system (10), and generating a position and / or orientation of the virtual coordinate system (10) by suitable computing means.

9. Computer-implemented method according to one of the preceding claims, wherein performing the method does not require registration fiducials attached to the patient.

10. Computer-implemented method according to one of the preceding claims comprising at least one of the following steps: displaying the virtual coordinate system (10) in the image data, displaying the tracked instrument in the image data, displaying thepatient sensor (20, 30) in the image data, and displaying a surgical target in the image data.

11. Method for surgical navigation comprising the steps of fixating a sensor in a field of view of a surgical navigation system, in particular on or near a patient, detecting a spatial position of a tracked instrument in the tracker coordinate system, and executing the computer-implemented method according to one of the preceding claims.

12. A computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method of any one of claims 1 to 10.

13. A computer-readable medium comprising instructions which, when executed by a computer, cause the computer to carry out the method of any one of claims 1 to 10.

14. A surgical navigation system (1) comprising means for carrying out the method of any one of claims 1 to 10.

Images