Stereotactic computer assisted surgery based on three-dimensional visualization

ES 2 595 366 T3

DESCRIPTION

Computer-aided stereotactic surgery system based on three-dimensional visualization.

BACKGROUND OF THE INVENTION

The invention relates to a computer-assisted surgery (or CAS) system that uses stereotactic navigation with a three-dimensional display, and, more specifically, to a CAS system that is reactive in nature and does not interfere with workflow procedures in the operating room.

A current procedure for inserting implants (consisting, for example, of a plate and its corresponding screws) is usually carried out by positioning the plate in the corresponding anatomical location and inserting the screws with the aid of fluoroscopy. Implantation of pin and plate systems is often challenging because operating room procedures are generally minimally invasive and therefore placement is performed by trial and error with fluoroscopy, for example with a C-arm apparatus, that is, C-arm vision. Generally, the result is long operating times. In addition, during this procedure, both the patient and the surgeon are exposed to significant amounts of radiation.

Furthermore, in some cases it may be impossible to determine the position of the implant components (eg screws in the bone) with sufficient precision, because the fluoroscopic image is only a two-dimensional image; This can lead to incorrect placement or the insertion of screws of inadequate length. In turn, this can lead to high revision rates or even injury (eg, hip joint injury). Therefore, to ensure that these implant components do not protrude from the bone, it is sometimes necessary to position these implant components leaving a gap from the edge of the bone with an excessively large margin of error. As a consequence, in many cases the implant cannot be positioned as intended, and the desired biomechanical stability cannot be achieved. In the case of femoral neck fractures, for example, no significant improvement is obtained by conventional fluoroscopic navigation.

Other leading-edge technologies currently being used in operating rooms to assist surgery include three-dimensional (3D) imaging and intraoperative navigation systems based on tracking technology. However, these technologies are only used in a few hospitals. The low implementation of these technologies is mainly due to their high cost, the effort involved in installing these systems and the important changes introduced in procedures or workflow in the operating room. For example, tracking technologies require that there be a line of sight between the tracking device and the navigation detection system; This interferes with normal workflow as the surgeon and other surgical personnel must always keep in mind the system requirements for line of sight. Furthermore, in general, successful positioning of a primary implant, such as a plate or nail, cannot be defined in the preoperative phase. For example, during an operation, positioning can be accomplished by haptic matching on the bone surface or by reaming the bone to create a space for an intramedullary nail. Furthermore, although the position of the subimplant (s) can be based solely on preoperative images (eg. g., fluoroscopy or CT images), this position remains relative to the position of the main implant. Therefore, a positioning procedure cannot be fully planned in the preoperative phase, but must be optimized during the operation. In this regard, classical stereotaxy cannot be used due to the fact that it is not possible to predefine the position.

Accordingly, there is a need for a computer-assisted surgery (CAS) system that improves surgical procedures without significantly interfering with workflow in the operating room. More specifically, there is a need for a combined CAS and 3D imaging system that can be easily integrated into the clinical environment. Preferably, such a system would be low cost, easy to install and use, and would minimize changes to the workflow in the operating room.

In EP 1 523 950 A1, a method and apparatus for percutaneous or minimally invasive implantation of a construct, particularly in the spine of a patient, is described. The construct can be implemented using a navigation system to plan and execute a non-image-based procedure. It is possible to connect a plurality of parts of the construction together by the determined positions and trajectories and navigate with the navigation system. Instead of fluoroscopic imaging, electromagnetic localization and tracking elements are used for navigation during implantation of parts of the tube.

ES 2 595 366 T3 construction.

RESUME

In a reactive procedure for stereotactic surgery, the procedure preferably comprises positioning an implant associated with a reference body in a region of interest of a patient's anatomy; the detection of information related to the implant by means of an imaging system; the determination, based on the information detected regarding the implant, of an action to be carried out as part of the surgery; and the visual presentation of the information on the position associated with the implant and the area of ​​interest, depending on the action to be carried out.

Positioning may comprise the acquisition of two fluoroscopic images of the area of ​​interest at two different angles.

The display may also comprise the processing of the detected information regarding the implant by estimating the contours of the area of ​​interest in at least two dimensions, based on the plurality of two-dimensional images.

Detection may comprise detecting the presence of the reference body, based on one or more reference markers.

The method may preferably comprise positioning a medical device associated with a reference body proximate to an area of ​​interest of a portion of a subject's anatomy and imaging the area of ​​interest at two or more angles to obtain a plurality of two-dimensional images. In a preferred embodiment, the reference body comprises a plurality of reference elements, more preferably at least four of these markers that are visible to the imaging system. It is also preferred that the reference markers comprise spheres that are visible to the imaging system.

The plurality of two-dimensional images can be processed to produce three-dimensional information associated with the area of ​​interest. Furthermore, the method may preferably also include the association, based on the three-dimensional information, of a virtual medical device with the area of ​​interest and the reference body and the visual presentation of the association as an image showing the superimposed virtual medical device. over the area of ​​interest.

The virtual medical device can comprise a main implant and one or more sub-implants. In addition, the virtual main implant can be superimposed on the current location of the real implant and the virtual sub-implants can be generated in order to show its future position. Therefore, virtual subimplants can inform the surgeon of where the real subimplant will be located before placing it in the area of ​​interest.

Imaging preferably comprises acquiring two fluoroscopic images of the area of ​​interest at two different angles. Furthermore, the processing may also preferably include estimating the contours of the area of ​​interest in at least two dimensions, based on the plurality of two-dimensional images.

Processing may also comprise forming a three-dimensional image associated with the area of ​​interest, based on the estimate. The procedure can be applied to a surgical implant procedure in which the area of ​​interest comprises a femoral head. The plurality of two-dimensional images may comprise anteroposterior and axial images of the femoral area and the estimation may comprise profiling of the femoral head on the anteroposterior and axial images. In this regard, the method may also comprise forming parts of a three-dimensional sphere representing prominent parts of the femoral head.

As another possibility, the medical device may comprise an intracapsular plate and the reference body may be connected to the plate, and the positioning may comprise positioning the intracapsular plate in a femur, near the femoral head. Furthermore, the virtual medical device may preferably comprise a virtual intracapsular plate and the display comprises the presentation of the virtual intracapsular plate superimposed on the position of the intracapsular plate relative to the femoral head.

ES 2 595 366 T3

The present invention is a computer-assisted surgical system, as defined in claim 1. Preferably, the reference body can be detected and superimposed on an object that models the area of ​​interest, e.g. eg, a ball for a femoral head; and then the display signal can be generated.

The processor can process the one or more two-dimensional images by tracing the contours of the area of ​​interest in two dimensions and creates a three-dimensional object that represents the area of ​​interest. The three-dimensional object can be generated from a database, depending on the age and sex of the patient. The three-dimensional object can also be determined based on certain characteristic elements associated with the area of ​​interest.

A medical device may comprise a device selected from the group consisting of an intracapsular plate, an artificial joint, a pacemaker, and a valve.

Another aspect may be a stereotactic navigation computer-assisted surgery (CAS) system with three-dimensional visualization, in which an implant or implant system acts as a stereotactic device. A reactive CAS system designed for use with mono and polyaxial plates and nails can be provided. In accordance with the principles of stereotaxy and 2D-3D runs, a system can be provided that virtually suggests or signals the optimal position of an implant by calculating that position. In addition, the system can also calculate screw lengths before drilling. Assisted by 3D virtual visualization and image processing, the system can achieve optimal biomechanics.

Also, unlike existing navigation systems, the CAS system can be designed to be reactive, in order to save the surgeon any extra effort. In particular, the system can be activated through the use of a reference body, Kirschner wire implants, or screws that are normally used as part of the surgical procedure. In addition, by detecting these devices, the system is able to determine the stage of the workflow that is being carried out. More specifically, image processing is used to detect various objects during the workflow and determine what stage the surgeon is performing and for system adaptation.

In another aspect, the system can provide the necessary 3D information without requiring an intraoperative 3D imaging system (eg, 3D C-arms). The system can also be low cost, easy to install and use, and can minimize changes to the workflow in the OR. The present system may also require fewer X-rays and therefore may be safer for patients.

In another aspect, a procedure may be performed iteratively (eg, the use of PCI to repair a femoral neck fracture) that includes one or more of the following steps:

1. positioning of an implant in an anatomical area of ​​interest, eg. eg, based on satisfactory haptic correspondence;

two. fluoroscopic imaging of the anatomical area of ​​interest;

3. virtual checking of the future position of the subimplant (or subimplants);

Four. virtual realignment of the implant according to the constraints until a satisfactory virtual position is reached;

5. provision of active or passive realignment values ​​of the implant position for the surgeon (that is, actively, identifying the best location; or passively, leaving it to the surgeon to decide);

6. actual realignment of the plate by the surgeon, based on realignment values ​​and satisfactory haptic correspondence; Y

7. iteration of the procedure from stage 2 to completion of the operation.

These and other aspects and features are described in more detail below.

BRIEF DESCRIPTION OF THE DRAWINGS

ES 2 595 366 T3

Fig. 1A illustrates a stereotactic computer-assisted surgery system.

Fig. 1B shows a computer that can be used in the system of FIG. 1.

Fig. 2 illustratively shows the possibility that a conventional two-dimensional (2D) image does not accurately indicate screw positions in an area of ​​interest.

Fig. 3 illustratively shows the possibility of using three-dimensional imaging techniques to present positional information that is not revealed by conventional two-dimensional images.

Fig. 4A is a flow chart illustrating a procedure for implanting a medical device.

Fig. 4B is a flow chart illustrating a procedure for positioning an implant.

Fig. 4C is a flow chart illustrating a procedure for generating a virtual image of an implant and a site of interest.

Fig. 5 shows the placement of an intracapsular plate implant.

Fig. 6A is a side view of an implant system that includes a reference body and an implant.

Fig. 6B is a perspective view of a reference body and an implant.

Fig. 6C is a perspective view of a reference body and an implant.

Fig. 7 illustrates the placement of an intracapsular plate implant in a femur.

Fig. 8 illustratively shows the stage of taking two fluorographs captured from different angles.

Fig. 9 illustrates the detection of the femoral head in the two fluorographs.

Fig. 10 illustrates the visualization of a virtual three-dimensional sphere representing the femoral head as a function of conical projections of the two two-dimensional fluorographs.

Fig. 11 shows the stage of presenting a visualization based on a correspondence of the three-dimensional sphere with the two-dimensional images.

Fig. 12A shows a step of automatic adjustment of the proposed position of the intracapsular plate in the distal direction.

Fig. 12B shows a step of automatic adjustment of the proposed position of the intracapsular plate in the distal direction.

The fig. 13A shows a step of automatic adjustment of the proposed position of the intracapsular plate by external rotation.

Fig. 13B shows a step of automatic adjustment of the proposed position of the intracapsular plate by external rotation.

Fig. 14 shows the repositioning and fixation of the intracapsular plate as a function of the proposed position.

Fig. 15 shows automatic Kirschner wire insertion detection and femoral head movement detection to ensure reactive behavior.

Figs. 16A through 16H illustrate the use of the method of FIG. Four.

DETAILED DESCRIPTION

In general, in one aspect, the system of the present invention is based on fluoroscopic image registration.

ES 2 595 366 T3 with an implant associated with a reference body. For example, the implant (eg, a stable angle plate) may include the reference body or be positioned at a predefined location relative to the reference body, which is detected or recorded on a fluoroscopic image. In this way, the actual position and spatial dimension of the implant can be determined by correctly identifying and registering the reference body on fluoroscopic images.

When multiple implants are included as part of the procedure, e.g. For example, major implants and sub-implants, following the registration of the main implant described above, the location of any of the remaining sub-implants can be virtually illustrated in the correct spatial position relative to the fluoroscopic images of the main implant. Subimplants (p. (e.g., associated stable angle plate screws) will be located in a predefined fixed position relative to the main implant after all implants have been implanted.

In order to provide the information necessary for an anatomically correct location of all implants (main and sub-implants), an approximation to the important anatomical areas is made by means of bodies or three-dimensional objects illustrated in the fluoroscopic image in a correct relative position. The desired values ​​are compared to the location values ​​of the rest of the implants, which are used to determine the current position of the main implant. During preoperative planning (for example, using a non-invasively applied reference body), partial or sub-implant implants (e.g. screws) can first be placed in an optimal position, regardless of the location of the main implant ( license plate). In a subsequent operation (using an invasive reference body), where the location of the main implant has been determined by preoperative planning (with an estimate of the position made by the surgeon), the location of the main implant can be optimized by haptic feedback. After the registration described above, the resulting sub-implant or partial implant location is displayed virtually; this position is compared with the position of the partial implant in the preoperative plan and with the distances to important anatomical (three-dimensional) structures. In a reactive iterative process (plate fit per system instructions), it is possible to determine the optimal balance between an ideal main implant location (e.g. plate fit) and the ideal partial implant position (e.g. the location of the screws).

Referring now to fig. 1A, a computer-assisted stereotactic surgical (CAS) system 100 is illustrated. As shown, in the embodiment, system 100 includes imaging apparatus 110, such as a C-arm fluoroscope, and a computer 120, such as a laptop. In general, computer 120 contains a processor 150, memory 160, and other components normally present in general-purpose computers, as illustrated in FIG. 1 B.

Information that can be accessed by processor 150 is stored in memory 160, for example through a bus 162, including instructions 164 to be executed by processor 150 and data 166 that is retrieved, manipulated, or stored by processor 150. Memory 160 can be of any type capable of storing information that processor 150 can access, such as a hard drive, ROM, RAM, CDROM, writable, read-only, or the like. Instructions 164 may comprise any set of instructions executable by the processor directly (eg, machine code) or indirectly (eg, scripts or scripts). In this regard, the terms "instructions", "steps" and "programs" may be used interchangeably herein. The functions, procedures and routines of the program are explained in more detail below.

Data 166 can be retrieved, stored, or modified by processor 150 in accordance with instructions 164. The data can be stored as a collection of data. For example, the data can be stored in computer records, in a relational database as a table with a plurality of different fields and records, or in the form of an XML document. The data can also be formatted to give it any machine-readable format, such as, but not limited to, binary values, ASCII, or EBCDIC (Extended Binary Coded Decimal Interchange Code). Furthermore, along with the data any information that is sufficient to identify the relevant data can be stored, for example descriptive text, own codes, pointers or information used by a function to calculate the relevant data.

Although processor 150 and memory 160 are functionally illustrated in FIG. 1B within the same block, those skilled in the art will understand that processor 150 and memory 160 may actually comprise multiple processors and memories that may or may not be arranged in the same physical housing. For example, some or all of the instructions 164 and data 166 can be stored on removable CD-ROMs, and others on a chip.

ES 2 595 366 T3 read-only computer. In addition, some or all of the instructions 164 and data 166 may be stored physically remote from the processor 150, yet still accessible by the processor. Similarly, processor 150 may actually comprise a collection of processors that may or may not work in parallel.

As shown, a computer 120 may comprise other components normally found in a computer system, such as a screen (eg, an LCD monitor), user input (eg, a keyboard, mouse , gamepad or touch screen), microphone, modem (eg, phone or cable modem), and all the components used to connect these items together.

As also shown in fig. 1A, a patient 185 would normally be positioned on an operating table with various restrictions, such that movement of the area to be operated on is restricted during surgery. Fluoroscope 110 (or other suitable imaging apparatus) is used to image the area of ​​interest from the patient's anatomy, e.g. For example, the area to be operated on or an area where the implant will be fixed. As explained in greater detail below, an illustrative zone of interest may comprise a zone that includes the femoral neck and an intracapsular plate (ICP). Computer 120 (or other suitable imaging and processing apparatus) is used to process fluoroscope images, determine implant and subimplant positioning, and provide feedback / instructions to the surgeon. The processing steps carried out by the computer are described below.

In another aspect, a problem is addressed with the current ICP implantation technique in the precise positioning of the plate using two-dimensional (2D) images. This problem is due, in part, to the danger posed by the placement of screws necessary to avoid recesses. Specifically, it is necessary that the ends / tips of the screws are arranged as close as possible to the second shell. However, the 2D images used by the surgeon do not reflect the three-dimensional nature of the problem.

In fig. 2, some common drawbacks of a conventional two-dimensional (2D) image are shown. In particular, a 2D image 200 may not indicate improper screw positioning. In this case, the 2D image 200 makes it appear that the screws are positioned correctly within the bone. However, a 3D illustration could provide other information showing that a screw may have actually pierced the bone. For example, fig. 3 illustrates how 3D imaging can reveal a screw positioning problem that is not apparent with conventional 2D imaging. In fig. 3, none of the 2D images 300, 310 show any problem with the position of the screws. However, if the 2D images are combined to create a 3D visualization, it becomes apparent that the screw tip protrudes through the bone, as illustrated at 320. Therefore, the 2D images currently relied on by surgeons may not always accurately reflect the location and positioning of medical devices and the like within an area of ​​interest. Therefore, in this example, the ability to access 3D images would be beneficial to the surgeon.

In one aspect, a system and method can be provided that generate 3D information from 2D images to allow more precise positioning of a medical device, e.g. g., an implant, and thus avoid the aforementioned problems. Generally, as used herein, the term "medical device" includes any biomedical device or structure that is introduced or implanted into the anatomy of a subject. Such devices include those that replace or perform the function of absent biological structures, or that are placed on top of or within bones or parts of the anatomy. As mentioned above, the illustrative example of intracapsular plate (ICP) implantation can be used to repair a femoral neck fracture. Note, however, that the system and procedure described may have application in numerous classes of surgery, including virtually all fields of bone surgery (eg, trauma, orthopedics, and pediatrics).

By way of background, it is generally known that fractures are usually repaired by reduction and fixation of fractured bones. The individual bone fragments are aligned in their normal anatomical position (ie, reduced) so that the separated parts can be reattached with the growth. It is necessary that the parts remain relatively stable with respect to each other over a long period of time to allow healing. In some cases, particularly for more complicated fractures, it is necessary to connect the fractured pieces of bone directly to each other. In these cases, the fracture is fixed or reduced by an invasive procedure in which an implant is installed inside the body with screws or nails.

ES 2 595 366 T3

Referring now to fig. 4A ,, a high-level flow chart 400 of the stages of the implant implantation procedure is illustrated. As shown, the procedure begins with the positioning of a primary implant in an area of ​​interest, in step S402. As explained in more detail below, this initial positioning is preferably performed using fluorographs taken in at least two dimensions or directions. Once the surgeon considers the position of the main implant satisfactory, the system 100 generates an image showing the position of a virtual implant and the corresponding virtual subimplants with respect to the area of ​​interest, step S408, based on the fluorographs and the position. of a reference body or reference objects within the field of view of the fluoroscope 110.

By imaging the virtual implants, the surgeon can then fix the implant, by sub-implants, for example, as illustrated in S424. Once the subimplants (eg screws) and implants are in place, the system can perform a quality check, at S428, by detecting and displaying the actual location of these implants relative to their desired position. This quality check is desirable, since during implantation, the position of an implant or subimplant may change from its ideal position due to mechanical forces exerted during, for example, drilling or screw placement or as a result of movements of the implant. patient. In this regard, quality checks can also be carried out, as in step S428, during implant fixation, in step S424. Furthermore, postoperative quality checks can also be carried out by the system to detect movements in the implant caused, for example, by patient activity.

Significantly, the above procedure 400 is reactive in nature, as the surgeon does not need to inform the system 100 of what step he or she is performing as part of the operating room workflow. In this regard, this system is compatible with normal operating room workflow and is capable of determining the stage of operating room workflow that is being carried out, for example by detecting the presence of a reference body or object.

Referring now to fig. 4B, the sub-steps or the procedure for positioning or aligning the implant according to step S402 of FIG. 4A. As shown, the procedure begins with the insertion and positioning of the main implant in the anatomical area of ​​interest in step S430. To continue with the illustrative example, an intracapsular plate (ICP) is used to repair femoral neck fractures. Thus, at step S430, the ICP plate would be inserted into the patient and roughly positioned on the bone, in this case the femoral neck. This step can be carried out, for example, in accordance with normal workflow in the operating room, for example allowing the physician to use haptic feedback to judge an appropriate initial position of the plate.

In this regard, fig. 5 illustrates the placement of an ICP 510 implant (eg, the main implant) in conjunction with subimplants (ie, the screws) to secure a femoral neck fracture. As shown, the ICP 510 is attached to the femur 520 and screws are inserted through the neck and head of the ICP. Preferably, the screws entering the femoral head are fully positioned within the head. In yet another aspect, since the ICP is adapted to the shape of the femur, the degrees of freedom in positioning the ICP on the bone are limited and are used as part of the S402 alignment procedure. Specifically, the ICP can only be displaced along (ie, translation) and / or rotated around the axis of the femur. Furthermore, the ICP is provided with threaded holes, so that the position / angle of the screws with respect to the plate is known.

Prior to insertion of the main implant into the area of ​​interest, the main implant 520 is connected to a reference body or object. The reference body is attached to (or is part of) the implant, but can also be attached to a guiding device or instrument (eg, a drill guide). In this way, the position of the implant can be determined based on the location and position of the reference body. Preferably, each implant is associated with a different reference body that is detectable by system 100, in particular fluoroscope 110. In a preferred embodiment, the reference body comprises a plurality of spherical reference markers inserted above or within the instrument (eg guiding device). By arranging the reference markers in a predetermined pattern, they can serve as identifiers for the different instruments. In this regard, the markers and the reference instrument may conveniently be referred to as a reference body, although it is the reference markers that provide the reference.

For example, fig. 6A illustrates a side view of a reference body 604 as part of the implant 610. Together, the reference body and the implant are referred to herein as the implant system 614. As shown in FIG. 6A, reference body 604 includes one or more reference markers 616 that

ES 2 595 366 T3 are detected by the imaging system and used as reference or measurement points. Preferably, the reference markers comprise spheres to make detection easier within a two-dimensional imaging system such as a fluoroscope. Furthermore, the arrangement of the reference markers within the reference body acts as a signature that is used to identify the reference body and the associated implant. Since the dimensions of the reference body and the implant are known and these devices are fixed relative to each other, the location of the implant can be accurately determined by detecting or recording the location of the reference body. As also shown in fig. 6A, reference markers can also be placed on the implant itself, but are not required.

Fig. 6B is a perspective view of guiding device 604 and implant 610 (which, to continue the example, is an ICP) in a separate state. Fig. 6C shows these two devices in a joined state. As shown, the guiding device 604 has a suitable contour to fit the ICP 610. In addition, it includes openings that allow access to the screw holes of the ICP 610 that are used to secure the implant 610 as explained in greater detail below. To allow processing, the reference body must be located in the field of view of the image with the implant and the area of ​​interest 640, which, to continue with this illustrative example, includes the neck of the femur and the femoral head. As part of this initial insertion and placement, the surgeon will typically use haptic feedback to determine a starting location for the implant.

Fig. 7 shows an instrument 700 that can be used in the initial insertion and placement of the ICP 610 on a femur 720. In addition, when performing this initial placement, the surgeon can use either the instrument 700 or the reference body 604. A In this regard, instrument 700 may also comprise a reference body by placing appropriate reference markers on top of or within it.

Returning to fig. 4B, once the surgeon determines an initial location for the implant (and the reference body that accompanies it), a fluorography of an area of ​​interest is taken in a first dimension or direction, in step 434. For example, You can take a fluorography in the anteroposterior dimension or the axial dimension. With reference to fig. 1, the anteroposterior view is illustrated with the emitter 190 and detector 192 aligned along the y-axis, while, in the axial view, the emitter and detector are aligned along the z-axis. As shown in fig. 8, fluorographs can be taken from any two different dimensions or directions 810A, 810B. Preferably, the two images will be taken perpendicular to each other (that is, forming an angle close to 90 degrees between them), but this is not essential and any angle will suffice.

Because the implant and the reference body are located within the visual field of the imaging device, and because of their proximity to the area of ​​interest, the fluoroscope 110 detects the presence of the reference body, that is, the reference markers. Computer 120 then uses the image data it receives from fluoroscope 110 to provide a visualization of the implant location relative to the area of ​​interest. In particular, the registration of fluoroscopic images is carried out through the reference body. As explained above, the reference body is normally in a fixed position with respect to the implant and the bone. In addition, a disk is used in the image intensifier to develop the X-ray image. The disk is usually attached directly to the image intensifier. Such a reference body would also be observed in each image and could be used to compensate for image distortion and for determining the center of, for example, the X-ray beam.

Determination of the implant with respect to the anatomical area of ​​interest is performed by known image processing techniques based on the variation in spatial radiation reaching the detector, including radiation directed towards the area of ​​interest and the reference body. Through spatial variation, the computer is able to construct an image that accurately represents the spatial relationship between the implant and the area of ​​interest (p. g., femur and femoral head) as a two-dimensional image.

After observing this image, the surgeon can then determine if the implant should be repositioned, as in step S438. For example, the surgeon may decide to adjust the position along the femur by bringing it closer to the femoral head or another degree of freedom. If the surgeon decides that such adjustment is warranted, he repositiones the implant as shown in step S440 and more fluorographs are taken in step S434. On the other hand, if the surgeon determines that no adjustment is necessary in this dimension, the procedure continues at step S442 with stabilization of the implant. To continue with the example, stabilization could be accomplished by inserting a Kirschner wire (K-wire) through one or more openings in the ICP.

With the implant attached as described above, a one-dimensional fluorography can be taken

ES 2 595 366 T3 different, step S446. In particular, if the fluorographs in step S434 were taken in the anteroposterior direction, in step s446 they can be taken in the axial direction or at another angle. In this regard, as part of step S402, it may be sufficient to use a single image for this step in order to optimize the position in only one degree of freedom (eg, a distal implant displacement) for which no 3D information is required.

After completing the fluorography taking in step S446, the surgeon can then view an image of the implant position. If it is determined that the implant needs to be adjusted in step S448, e.g. For example, rotate it in the case of an ICP, the procedure returns to step S446 and more fluorographs are taken in this dimension. Once the surgeon is satisfied that the implant position is adequate based on images obtained in this dimension, the procedure continues at step s450 with further stabilization of the implant. For example, when the implant or medical device is an ICP, K-needles can be inserted through additional openings in the ICP. As a result of the above procedure, a surgeon can determine the position of the ICP or other implant iteratively and in accordance with normal operating room workflow procedures. That is, the surgeon can repeat any stage within the procedure until the implant is correctly positioned.

With the implant positioned as described above with respect to step S402, the procedure then continues as shown in step S408 of FIG. 4A and as will now be described in more detail, as shown in FIG. 4C. In particular, at step S454, the system can then generate 3D information from the two-dimensional fluorographs recorded at step S402 or two other two-dimensional fluorographs can be taken at different angles, as described above. Since the implant is now stabilized with respect to the area of ​​interest, further fluorography can be taken with the K-needles acting as a trigger for the system. Furthermore, since the reference body can also be attached to the implant, it can also serve as a reference object as described above.

The resulting 2D images are processed to locate and trace a three-dimensional outline, that is, a sphere, of the femoral head. For example, fig. 9 shows an anteroposterior (AP) view image 910 and an axial view image 920 with overlapping circles 930, 940 tracing the contours of the femoral head. Circles 930, 940 can be constructed by computer 120 through image processing techniques such as edge detection or computer generated patterns. Such models can be created preoperatively using MRI or other non-invasive techniques that can determine the location and size of organs or bones within the area of ​​interest.

Furthermore, by means of the 2D images, the computer then determines and generates a 3D object that is associated with the area of ​​interest and models it, step S456. In particular, fig. 10 illustrates the visualization of a virtual 3D sphere representing the femoral head, based on conical projections of 2D images 910, 920. As shown in FIG. 10, the virtual 3D sphere is formed by projecting the two-dimensional coordinate system onto a three-dimensional coordinate system. In this example, because the profile of the femoral head forms a circle, projection onto a three-dimensional coordinate system results in a sphere. Depending on the contours of the area of ​​interest, these projections can be carried out using a Cartesian and / or spherical coordinate system. Furthermore, the location of the object with respect to the area of ​​interest can be precisely determined based on the position of the implant in relation to the reference body.

Fig. 11 shows the stage of presenting a visualization of the area of ​​interest, implants and sub-implants, based on the establishment of a correspondence of the 3D sphere with the 2D images, step S458. As shown in fig. 11A, a virtual ICP 1104 presents screws and a spherical profile of the femoral head superimposed on the original axial and AP 2D images. This visualization allows the surgeon to easily observe the position of the ICP and screws relative to the femoral head. In particular, the visualization shows the virtual screw positions, their length, and how they will be positioned within the femoral head. Also, the system may suggest the screw, p. For example, the length or specific model of the screw, which would be adequate to fix the implant.

Visualization also allows the surgeon to manually adjust the actual position of the ICP if better alignment is deemed necessary. For example, in fig. 12A shows a proposed adjustment of ICP 510 in the distal direction, as seen from the two-dimensional AP image. More specifically, as can also be seen in fig. 12B, the display may also include areas that indicate a preferred translational fit of the implant relative to its current location. For example, an acceptance zone 1220 (eg, by colors) can be used to indicate a more preferable location. Thus, if the screws are located distally out of

In this area, the surgeon can manually adjust the plate at S55 and observe an updated display by returning to step S52.

In figs. 13A and 13B show how the adjustment of the ICP 610 can be achieved by external rotation 1300. In particular, similar to what occurs in the case of translational adjustment, if the surgeon believes that the implant is not aligned correctly , you can adjust the 1300 arrows to visualize how the implant aligns when rotated. As can also be seen in fig. 13B, an acceptance zone 1340 can be used to show how rotation would change the location of the screws relative to a preferred position. As also shown in fig. 13B, the rotational movement is preferably carried out after the location along the neck of the femur has been satisfactorily determined. In this way, the reference body 604 can be fixed to the bone by a nail 1326. This nail 1326 can be used with a sprocket mechanism 1330 to more precisely rotate the implant as shown.

As explained above, the system can be considered reactive, as it reacts to the surgeon rather than asking him to perform an action or interact with the computer or system. Thus, if the surgeon decides that the implant is correctly aligned, then he can decide to secure the implant and conclude the procedure; This minimizes interference in the current workflow in the operating room and allows the surgeon to use their own judgment as part of the workflow. In contrast to this, conventional solutions tend to interfere with the workflow in the operating room, since they ask the surgeon to interact with the CAS; This lengthens the surgical procedure and is more demanding in terms of equipment, thus increasing the cost of surgical procedures.

After completing the steps discussed above in relation to step S408, the procedure continues to step S424, in which the implant can be attached to the area of ​​interest. Further fluorographs can be taken in the course of or after step S424 to check the reduction of the fracture and the position of the ICP and K-wires or screws. For example, Figure 15 shows the insertion of K-wires 1620, 1630 and a detected movement of the femoral head 1610. Ideally, the system will detect such movement and propose a corrective measure, such as new screw lengths. If necessary, another implant repositioning and fracture reduction can be performed as explained above.

As explained above, Kirschner wires (K-wires) can be inserted through openings in reference body 604. More specifically, as shown in FIG. 15, a first K 1630 needle can be inserted to fix the plate to the bone. (A second K 1620 needle can also be inserted through the fracture.) Screws can then be inserted to compress the fracture (S424). The screws can be self-tapping or can be inserted through drilled holes. An edge detection based image processing computer application is preferably provided to detect any insertion and bending of the screws or K needles. Such computer application may comprise a component or routine in a set of instructions that carries out the procedure. previously described. The ICP has threaded screw holes so that the position / angle of the screw relative to the plate is fixed. The K-wires can be removed before or after the screws have been inserted.

Another possibility is that the ICP is directly attached to a guidance device with a graduated scale in combination with an elongated hole (into which a K-needle can be inserted) that can help with assembly and any subsequent adjustments. that the surgeon deems necessary. In fig. 14 repositioning by prior art one shot application apparatus 1400 is shown; which is preferably replaced by the reference body 604 and other accessories 1326, 1330 discussed above.

Referring now to Figs. 16A through 16H, an alternate use of the procedures described above is shown. As will be explained in more detail, these figures show the insertion of a 1645 locking nail, which is used as part of a hip fracture repair system. In particular, and with reference to fig. 16A, the procedure begins with the insertion of nail 1654 into femur 1658. As also shown, nail 1654 is attached to a guiding instrument or device 1660. The guiding device is preferably provided with a plurality of reference markers, e.g. eg, four or more, which function as a reference body that is detectable by the imaging system. According to the procedures described above, in this initial stage of the procedure, the surgeon obtains fluorographs along a first dimension. For example, a fluorography can be obtained along the patient's anteroposterior axis or at any other angle the surgeon deems appropriate.

ES 2 595 366 T3

As shown in fig. 16B, if the image is taken in the anteroposterior direction, the computer 120 detects and calculates the position of the reference body and the instrument 1660, and presents a virtual nail 1666 relative to the area of ​​interest 1670. In addition, the display includes a projection 1674 of the location of a screw that will be used to secure nail 1654 within the femoral head or area of ​​interest 1670. As also illustrated in FIG. 16B, if the surgeon determines that the projected screw trajectory 1674 requires some adjustment, he may perform a translational adjustment 1678 of the nail into the femur 1658. After performing the translational adjustment 1678, the surgeon preferably takes another fluorogram. to confirm by display 1680, similar to that shown, that adjustment moved nail 1654 to a more desirable position.

Once the surgeon is satisfied with the translational fit of nail 1654, they can use the system to rotationally align the nail as shown in FIG. 16C. In particular, the surgeon would take a fluorogram at a different angle, such as in a lateral direction of the hip to obtain image 1684 shown in FIG. 16C. Using this image, the surgeon can rotate nail 1654 into a more desirable position and take more fluorographs to confirm the fit.

Once the surgeon determines that nail 1654 is properly aligned, he can insert a K-needle 1687 as shown in FIG. 16D. With the K-needle inserted, two or more two-dimensional images can be obtained with the fluoroscope, as described above. Using these two or more images, the system is able to determine the appropriate screw lengths, as shown in fig. 16E. In particular, two-dimensional images are used to create an object that models the area of ​​interest, in this case the femoral head. More specifically, when the area of ​​interest is the femoral head, the computer 120 uses these two-dimensional images to create a 1689 sphere and overlays within the sphere the locations and lengths of the screws that can be used to fix the 1654 nail. shown in fig. 16E, the display includes a virtual screw 1691 along with markings 1693 indicating the length of the screw that lies just inside the sphere 1689 and out through an opening in the nail 1654.

Based on the 1693 markings shown in fig. 16E, the surgeon can then select an appropriate screw of a desirable length to fix the nail 1654. Once the screw has been selected, it is inserted as shown in FIGS. 16F and 16G. As also described above, once a screw is in place, further images can be taken to verify that the length of the screw secures the device without protruding from the area of ​​interest as a result of the forces applied during the fixation procedure. , as illustrated in fig. 16H.

Image processing may include: detection and segmentation of anatomical features; detection of the position of the reference body; generation of 3D information from 2D images; and calculation of the optimal implant position. Furthermore, in another aspect, the system can propose an appropriate length for each screw.

As explained above, at least two 2D images containing the reference body may be necessary to provide 3D information. These images should be taken at different angles (preferably at an angle close to 90 degrees). Other 2D images can also be used to provide information. Images can be centered relative to each other by detecting distinctive anatomical features in the images and / or by the reference body. The reference body (which appears in every image) can be used to precisely center three-dimensional images. The reference body can also be useful in the automatic detection of these anatomical structures for segmentation (eg, detection of feature element boundaries). The relative position of specific anatomical structures to the reference body position can also be estimated based on statistics on the general shape of the bone and patient data (e.g. weight and height, gender, age ). This relative position can be used as a starting point for segmentation algorithms. Once the anatomical structures have been segmented, the image processing software application can correlate the structures of different images to generate 3D information.

Various three-dimensional reconstruction algorithms can be used to generate this information. Typically, algorithms will approximate segmented anatomical features using geometric shapes (eg, a circle). The geometric shapes are then correlated / centered with respect to their known relative positions in the 2D images. These shapes are then projected into 3D space to form, for example, a sphere or a cylinder. A typical 3D shape for an anatomical area can be initially selected from a database and mapped to the image by enlarging, rotating and / or translating the shape. The shape can also be altered, for example by an image transformation algorithm (morphing), to achieve a better match. In fact, they can

ES 2 595 366 T3 take preoperative images of the same anatomical area to better determine the true shape of various features.

Because the reference body is contained in each image and is attached to an anatomical area (eg, a bone), movement of the patient during surgery is not a problem. This is because the system can use the reference body location to center different fluoroscopic images (regardless of image content) and generate a true 3D image with few artifacts using 3D reconstruction algorithms. This aspect of precise image centering greatly reduces artifacts due to patient movement during surgery.

Preoperative planning can be carried out by taking preoperative images similar to intraoperative images. This preoperative planning can be used to determine optimal subimplant positioning, which can then be verified by comparing it to intraoperative positioning. Such preoperative images could be processed by different algorithms that would take too long to use during surgery, or could be manually segmented and correlated.

As explained above, a reactive workflow can also be provided by automatically detecting the status of an operation and thereby knowing the next operational steps to be carried out. In this way, suggestions can be offered to the surgeon. For example, a specific type, size, or shape of a best-fit implant may be suggested based on the detected geometry of a fracture. In addition, a previous suggestion can be modified based on additional information determined during surgery.

Other distinctive aspects may include the fact that the stereotactic device is implanted in the body. Additionally, 2D images (eg, fluoroscopic X-rays) can be used to generate 3D information. The Reference Plate (ICP) is contoured to match the contour of the bone surface to limit the degrees of freedom for adjustments. The reference plate (ICP) is also threaded to know the relative position of the screw. You can calculate and propose the position of the reference plate, the position of the sphere and the position and lengths of the screws.

Advantages may include the fact that it reduces surgery time for implant insertion, requires little interaction between the surgeon and the system, provides three-dimensional information about important areas, requires few changes in operating room procedures and it is cheaper than current tracking-based navigation.

Other features may include the fact that it takes into account any bending of the Kirschner wires (K-wires) through automatic detection, calculates and displays any dislocation of the femoral head during implantation, and calculates screw lengths.