Navigation system with automatic detection of tracking errors

Abstract

The present invention relates to a navigation system capable of automatically detecting possible tracking failures or errors during navigated surgery. In particular, the system comprises a tracking module and a topographic imaging module, each configured to acquire a different type of information from the surgical field, and a processor configured to check both sets of information and automatically detect tracking errors.

Description

[0001] NAVIGATION SYSTEM WITH AUTOMATIC TRACKING ERROR DETECTION

[0002] OBJECT OF THE INVENTION

[0003] The present invention is directed to a navigation system capable of automatically detecting potential tracking errors or failures during a navigated surgery. Specifically, the system comprises a tracking module and a topographic imaging module, each configured to acquire information of a different nature about the surgical field, and a processor configured to compare both pieces of information and automatically detect tracking errors.

[0004] BACKGROUND OF THE INVENTION

[0005] Surgical navigation systems allow the elements involved in an operation to be located at all times, including at least part of the patient's anatomy, so they can be displayed to surgeons in real time. To achieve this, it is necessary to ensure that if any of these elements shift from a previous position, the system has the means to measure this change and update its position correctly.

[0006] The usual solution in this field is for the navigation system to incorporate a tracking system, that is, a device capable of determining the three-dimensional pose in space of the patient's anatomy and / or other elements of the surgical field, such as surgical tools or instruments. The pose of an object is understood to be its position and orientation in space. The determination of the pose is always relative to a pre-established reference system.

[0007] The most widely used tracking systems are those based on an infrared stereo camera and individual fiducial markers in the form of reflective spheres (passive markers) or LEDs (Light Emitting Diodes, active markers), arranged in groups on rigid structures with known geometry. Such a tracking system visualizes the individual markers, determines their positions using triangulation, compares these positions with known geometries, and determines the pose of each navigated element (part of the patient's anatomy, surgical instrument, etc.). By associating a set of markers with a unique geometry with each element to be navigated, it is possible to update at least its position at all times. For example, to determine the pose of a part of the patient's anatomy when breathing, it is common to attach a set of markers to that part of the anatomy.In another example, particularly in spinal surgery, the patient marker assembly is typically a clamp or a small threaded rod (known as a Shanz pin) with fiducial markers attached, which are typically anchored either to one or more vertebrae (usually to the part known as the spinous or spinous process) or to the iliac crest. Since the fiducial marker assembly forms a single body with the element to which it is then attached, it can be kinematically ensured that the pose and movements of said marker assembly also determine the pose and movement of the element to which they are attached.

[0008] One of the biggest problems associated with this type of tracking system using fiducial markers is that unnoticed disturbances can occur in these markers, either in the individual markers, in the set of markers, or in their connection to the element to which they are attached. The usual way to detect these disturbances is either purely visual (the navigation system suddenly displays a clearly erroneous anatomical position compared to the actual surgical field) or through techniques that identify movement variations between the previous and subsequent measurements of the pose of each set of markers made by the tracking system.

[0009] These solutions are generally valid for significant disturbances, but other types of disturbances are at greater risk of going undetected. One example is a slow but continuous drift, perhaps caused by a slippage or rotation of the fiducial marker array attached to the patient's anatomy relative to its initial position. Another example is a small, punctual movement, which is macroscopically undetectable but impacts navigation accuracy and could lead to critical safety failures in surgeries requiring a high level of precision, such as laser cutting and implant guidance.Other less common but potentially safety-critical situations include failures in the monitoring system itself, such as an interruption or delay in communication that freezes the last valid measurement without alerting the user, or a single or cumulative defect in the accuracy of the measurement or its calibration parameters.

[0010] Other tracking systems, less common in the state of the art, are based on object tracking or point tracking algorithms, widely known in the field of artificial vision. These tracking systems comprise one or more cameras that acquire a plurality of images of the surgical field (in which a portion of the patient's anatomy is displayed and, optionally, other elements such as surgical instruments) and a processing unit capable of executing an object tracking algorithm. Said algorithm first selects characteristic points in a first image taken by one or more cameras. The selected characteristic points act as reference points that must be located, totally or partially, in the images taken by one or more cameras at subsequent time points, with the aim of inferring the movement of the surgical field relative to the measurement system.In this way, the tracking system tracks the three-dimensional pose of the surgical field (in particular, of a part of the patient's anatomy) by measuring variations in the position of reference points (generally translations and rotations) located in consecutive images.

[0011] By eliminating fiducial markers, the tracking system is simplified in terms of the physical elements required to perform its task, thus freeing up the surgical space, simplifying system assembly and configuration, and avoiding the mechanical disturbances to which the sets of markers relative to the patient's anatomy are subjected.

[0012] One of the problems associated with this type of tracking system is that the point tracking algorithm may not locate or may inaccurately locate the reference points in the image sequence, which would lead to erroneous estimations of the three-dimensional pose of the surgical field and, in particular, of the part of the patient's anatomy.

[0013] For example, if the surgical field is heavily affected by blood, fluids, or other undesirable elements during surgery, the tracking algorithm might not be able to properly locate landmarks. In an example where the tracking algorithm locates landmarks based on specific image characteristics (such as shadows or textures), which could be affected by various external factors such as lighting, landmark detection could be very inaccurate, and therefore, the 3D pose estimation would be erroneous.

[0014] Another potential source of error in these tracking systems is the cameras, which may fail to function at some point during surgery. In these cases, the tracking algorithm could continuously receive the same image, locating the reference points in the same positions and erroneously determining that the three-dimensional pose of the surgical field, and in particular of a portion of the patient's anatomy, is not changing over time.

[0015] The present invention proposes a solution to the aforementioned problems by means of a navigation system that, based on information from two modules of a different nature, is capable of detecting when a tracking failure has occurred and of emitting an alarm to alert the surgeon and medical staff about said tracking failure.

[0016] DESCRIPTION OF THE INVENTION

[0017] The present invention proposes a solution to the above problems by means of a navigation system according to claim 1. Preferred embodiments of the invention are defined in the dependent claims.

[0018] In a first inventive aspect, the invention provides a surgical navigation system comprising: a processor; a tracking module configured to acquire a three-dimensional pose of at least a portion of a patient's anatomy relative to a first reference system; a topographical imaging module configured to acquire at least one topographical image of at least a portion of the patient's anatomy, the at least one topographical image being referenced relative to a second reference system; the processor being in communication with the tracking module and the topographical imaging module, the processor being configured to establish a link relating the first reference system to the second reference system;or establishing a base reference system as well as a first link that relates the base reference system and the first reference system, and a second link that relates the base reference system and the second reference system; or carrying out in any order a) receiving the at least one topographic image from the topographic image module and establishing the three-dimensional pose of a region of the at least one topographic image with respect to the second reference system, and b) receiving, from the tracking module, the three-dimensional pose of at least a part of the patient's anatomy; or referencing the three-dimensional pose of the region of the at least one topographic image and the three-dimensional pose of the at least a part of the patient's anatomy with respect to the base reference system;and wherein the processor is further configured to: calculate a discrepancy measure between the three-dimensional pose of the at least one portion of the patient's anatomy acquired by the tracking module and the three-dimensional pose of the region of the at least one topographic image; and issue a tracking failure alarm if the discrepancy measure exceeds a predetermined value.

[0019] The system of the first inventive aspect comprises a processor, a tracking module and a topographic imaging module.

[0020] Throughout this document, a processor is defined as a processing unit configured to receive, store, and process data. This processor communicates with the tracking and topographic imaging modules, thereby exchanging data and commands between them.

[0021] The three-dimensional pose of an element comprises a set of parameters that allow the element to be positioned in space with six degrees of freedom. In particular, the three-dimensional pose comprises the element's three-dimensional position and orientation.

[0022] The tracking module is configured to acquire the three-dimensional pose of a portion of the patient's anatomy; that is, the three-dimensional position and orientation of said anatomy relative to a first reference system, integral with the tracking module and known to the processor.

[0023] In one embodiment, the tracking module comprises an infrared stereo camera and a plurality of fiducial markers, preferably in the form of reflective spheres and / or LEDs (Light Emitting Diodes), arranged in marker arrays of known geometries. These fiducial marker arrays are affixed to a portion of the patient's anatomy such that they are visible to the tracking module, which is configured to determine their positions by triangulation, compare said positions with geometries known to the tracking module, and determine the three-dimensional pose of the portion of the patient's anatomy to which they have been affixed. Optionally, fiducial marker arrays are also affixed to the topographic imaging module such that they are visible to the tracking module, which is configured to acquire the three-dimensional pose of the topographic imaging module over time.

[0024] Throughout the description, the three-dimensional pose of a portion of the patient's anatomy will be considered to be determined. When this portion of the patient's anatomy is part of the information manipulated by a processor, then the portion of the patient's anatomy is part of a numerical model or virtual model that allows the shape and pose of said portion of the patient's anatomy to be computationally determined.

[0025] In one embodiment, the tracking module comprises one or more cameras that acquire a plurality of images of the surgical field and a processing unit, which is part of the processor itself, capable of executing an object tracking algorithm, preferably the ORB-SLAM algorithm. Said algorithm selects a plurality of reference points in a first image of a portion of the patient's anatomy taken by one or more cameras and, subsequently, locates all or part of said reference points in the images taken by one or more cameras at subsequent time points. Thus, the tracking module tracks the portion of the patient's anatomy by measuring variations in the position of the reference points (generally, translations and rotations) located in consecutive images, and estimates the three-dimensional pose of said anatomy over time.

[0026] By eliminating fiducial markers, the tracking system is advantageously simplified in terms of the physical elements required to perform its task. This frees up the surgical space, simplifies system assembly and configuration, and avoids the mechanical disturbances to which the marker arrays relative to the patient's anatomy are subjected.

[0027] The topographic imaging module is configured to acquire at least one topographic image of at least a portion of the patient's anatomy relative to a second reference system, integral with the topographic imaging module and known to the processor.

[0028] Throughout the document, it will be understood that the topographic imaging module is configured to acquire measurements of a surface or a volume of at least part of the patient's anatomy and that the topographic image is a numerical model that represents said surface of the patient's anatomy or the surface of the volume of the patient's anatomy, the surface being referenced with respect to the second reference system.

[0029] In one embodiment, the topographic imaging module comprises a polarization-sensitive optical coherence tomography (PS-OCT) system and a processing unit that is part of the processor. The PS-OCT system takes measurements of a volume of a portion of the patient's anatomy, and the processing unit establishes the topographic image as the external surface delimiting the portion of the patient's anatomy comprised within the measured volume, determined by segmentation, thereby generating a numerical model representing said surface.

[0030] Examples of numerical models that represent surfaces are mathematical entities intended to represent either a surface or surface segments such as polynomial expressions or meshes where, in the latter case, a set of nodes and selected approximation models determine the points of the represented surface.

[0031] Another widely used example for representing a surface is a point cloud, where each point verifies that it is part of the represented surface. In this case, the point cloud includes a positional reference that allows the position and orientation of the set of points to be established. The translation and rotation operations on this point cloud are such that the same operation is applied to all points, resulting in a translation and / or rotation of the represented surface. When the represented surface behaves as a rigid solid, the relative distances between points in the cloud remain invariant. Particular cases where the use of point clouds is appropriate for representing surfaces are models obtained through PS-OCT and OCT measurements, since the data directly obtained after the segmentation operation of the volume on which the measurement was carried out is a point cloud.Based on this point cloud resulting from the measurement, it is possible to establish a selection that allows for more efficient manipulation.

[0032] In one embodiment, the topographic imaging module comprises an optical coherence tomography (OCT) system and a processing unit that is part of the processor. The OCT system takes measurements of a volume of a portion of the patient's anatomy, and the processing unit establishes the topographic image as the external surface delimiting the portion of the patient's anatomy comprised within the measured volume, determined by segmentation, thereby generating a numerical model that represents said surface. In one embodiment, the topographic imaging module comprises a stereo pair system and a processing unit that is part of the processor. The stereo pair system takes measurements of the surface of a portion of the patient's anatomy, and the processing unit establishes the topographic image as said surface, thereby generating a numerical model that represents the surface.

[0033] In one embodiment, the topographic imaging module comprises a structured light system and a processing unit that is part of the processor. The structured light system takes measurements of the surface of a portion of the patient's anatomy, and the processing unit establishes the topographic image as said surface, thereby generating a numerical model representing the surface.

[0034] In one embodiment, the topographic imaging module comprises an optoacoustic tomography system and a processing unit that is part of the processor. The optoacoustic tomography system takes measurements of a volume of a portion of the patient's anatomy, and the processing unit establishes the topographic image as the delimiting external surface of the portion of the patient's anatomy comprised within the measured volume, determined by segmentation, thereby generating a numerical model representing said surface.

[0035] The processor is configured to establish a link between the first reference system and the second reference system. Furthermore, it establishes a base reference system and the links between the first reference system and the second reference system with said base reference system.

[0036] Once this base reference system is established, the processor is configured to reference all data and information it receives against this base reference system.

[0037] The processor is further configured to receive the at least one topographic image from the topographic image module and to establish the three-dimensional pose of a region of the at least one topographic image relative to the second reference system. Furthermore, the processor is configured to reference said pose of the region of the at least one topographic image relative to the base reference system, since the processor has previously established a link between the second reference system and the base reference system.

[0038] Throughout the document, the region of at least one topographic image will be understood to correspond to the topographic image in its entirety or to a part of the topographic image.

[0039] The processor is further configured to receive the three-dimensional pose of at least a portion of the patient's anatomy from the tracking module, which is referenced relative to the first reference system, and to reference said three-dimensional pose of at least a portion of the patient's anatomy relative to the base reference system, since the processor has previously established a link between the first reference system and the base reference system.

[0040] Finally, the processor is configured to calculate a discrepancy measure between the three-dimensional pose of at least a portion of the patient's anatomy acquired by the tracking module and the three-dimensional pose of the region of the at least one topographic image. Both poses are referenced with respect to the base reference system, as mentioned above.

[0041] If the processor detects that the discrepancy exceeds a predetermined value, the processor issues a tracking failure (or tracking error) alarm to warn medical personnel that the images displayed by the navigation system may not correctly represent the current location of the surgical field.

[0042] Advantageously, the navigation system of the invention not only bases navigation on a tracking module, as is usually the case, but also has a second module, the topographic image module, capable of offering information of a different nature to be compared with the information from the tracking module. In this way, the navigation system is capable of detecting if the tracking module is not functioning properly, alerting medical personnel that the visual information displayed by the navigation system itself is not faithfully representing reality, as neither will other systems that are dependent on this information, such as a surgical system, which requires the position of the surgical field.Thus, if the tracking module is not accurately detecting the pose of a portion of the patient's anatomy (whether by means of markers or by means of object tracking algorithms), the navigation system of the first inventive aspect is capable of determining this inaccuracy thanks to the surface information provided by the topographic imaging module. In this way, although said inaccuracy is very subtle and may a priori go unnoticed, the navigation system of the invention is able to detect it, allowing, for example, the surgery to be temporarily interrupted to avoid causing unwanted damage to the patient. This functionality is especially relevant in surgeries that require a high level of precision, such as laser cutting and implant guidance surgery.

[0043] In one embodiment, at least a portion of the patient's anatomy is represented by a virtual model generated by the processor from a preoperative or intraoperative image, wherein the preoperative or intraoperative image is acquired by the topographic imaging module, by the tracking module, or by imaging equipment external to the system.

[0044] The virtual model can be generated from an image acquired before the start of surgery (preoperative) or from an image acquired during surgery itself (intraoperative).

[0045] These images can be:

[0046] An image acquired by the topographic imaging module before or during the operation;

[0047] An image acquired by the tracking module cameras before or during the operation; or

[0048] An image acquired by imaging equipment external to the navigation system, either preoperatively or intraoperatively. Examples of external imaging equipment that could acquire this type of image include an MRI scanner, or either 2D (also known as fluoroscopy) or 3D X-ray equipment, such as a CT scanner, or a C-arm or O-arm scanner.

[0049] In one embodiment, the at least one part of the patient's anatomy is a region that corresponds to the best approximation of the region of the at least one topographic image in the patient's anatomy, determined as a region that is verified to belong to the surface of the patient's anatomy and have the minimum distance to the region of the topographic image, the distance measured according to a pre-established norm, preferably the Euclidean norm.

[0050] During surgery, the surface of the patient's anatomy undergoes changes over time—for example, after a certain tissue is cut, after a fluid appears that obscures a portion of the patient's anatomy, etc.—and this is what we call a variable or temporally evolving surgical field.

[0051] This embodiment considers the surgical field to be variable, so the discrepancy measurement is established between specific regions of the topographic image and the patient's anatomy, whose pose is estimated by the tracking module. Preferably, these regions are regions that do not change or change minimally during a particular surgical interval, so that they are the ones that establish a reference region when the measurement is taken.

[0052] The topographic image region may be the entire topographic image or a portion or part of it. The processor may automatically select a portion or part of the topographic image as the region, or it may set the region as a portion or part of the topographic image previously selected manually by medical personnel.

[0053] The processor is configured to search for a portion of the patient's anatomy displayed by the tracking module that is closest to the topographic image region. This portion of the patient's anatomy must meet the following conditions: verify that it belongs to the surface of the patient's anatomy displayed by the tracking module; and be as close to the topographic image region as possible, the distance between regions measured according to a pre-established standard, preferably the Euclidean standard.

[0054] In this way, the discrepancy calculation is established between the poses of both regions; that is, the pose of the topographic image region and the pose of the part of the patient's anatomy that corresponds to the region of the patient's anatomy surface closest to the topographic image region. It is important to note that both surfaces involved in the pose comparison do not necessarily have to be equal, so it is necessary to establish the pose of one with respect to the other that provides the best degree of coincidence. This best degree of coincidence must be measured in some way and serves as a reference to establish the degree of discrepancy in the pose of one surface with respect to the other when both have a certain relative pose. The way to measure this degree of coincidence, according to various embodiments, is as described below.

[0055] In one embodiment, the discrepancy measure is calculated as:

[0056] - the difference between the three-dimensional pose of at least one part of the patient's anatomy at a given time instant and the pose it would have under the condition of showing the smallest distance with the three-dimensional pose of the region of the at least one topographic image at said time instant; or

[0057] - the difference between the three-dimensional pose of the region of the at least one topographic image at a given time instant and the pose it would have under the condition of showing the smallest distance from the three-dimensional pose of the at least one part of the patient's anatomy at said time instant; where the distance is measured according to a pre-established standard, preferably the Euclidean standard.

[0058] In this embodiment, the processor calculates the discrepancy based on the transformation that the three-dimensional pose of at least one part of the patient's anatomy would have to undergo to be as similar as possible to the three-dimensional pose of the region of the at least one topographic image, the poses having been acquired or estimated at the same time instant.

[0059] Alternatively, the processor calculates the discrepancy based on the transformation that the three-dimensional pose of the region of the at least one topographic image would have to undergo in order to be as similar as possible to the three-dimensional pose of at least one part of the patient's anatomy, the poses having been acquired or estimated at the same time instant.

[0060] In one embodiment, the at least one part of the patient's anatomy is a region resulting from segmenting the virtual model, preferably determining a boundary surface between different tissues, and wherein the discrepancy measure is calculated according to the following steps: - determining a set of points in the region of the at least one topographic image, or all of the points in the region of the at least one topographic image or a selection of representative points of the region of the at least one topographic image;

[0061] - for each point of the previous step, determining the point of at least one part of the patient's anatomy that is verified to have the smallest distance with said point of the region of the at least one topographic image;

[0062] - establishing the measure of the discrepancy between the three-dimensional pose of the at least one part of the patient's anatomy and the three-dimensional pose of the region of the at least one topographic image as a weighted measurement value of the set of distances between the pairs of points determined in the previous step, the point of the region of the topographic image and the point of the at least one part of the patient's anatomy.

[0063] In this embodiment, a specific way of calculating the discrepancy between the three-dimensional poses of at least one part of the patient's anatomy and the region of at least one topographic image is contemplated in which a direct comparison is not established between poses but between regions positioned according to said poses, which indirectly quantifies the discrepancy that exists between said poses.

[0064] To do this, the processor first determines a set of points in the region of the at least one topographic image. In a more specific embodiment, this set comprises all the points in the region of the at least one topographic image. In another more specific embodiment, this set comprises a selection of points representative of the region of the at least one topographic image.

[0065] For each point determined by the processor in the region of the at least one topographic image, the processor determines a point in at least one part of the patient's anatomy that is verified to have the shortest distance from the point in the region of the at least one topographic image. The distance between points is measured according to a pre-established standard, preferably the Euclidean standard.

[0066] In this embodiment, the portion of the patient's anatomy is represented by a segmented region of the virtual model. Preferably, this segmented region comprises a surface that is the boundary between two tissue types; for example, soft tissue and bone. Finally, the processor calculates the discrepancy between the three-dimensional poses of at least one portion of the patient's anatomy and the region of the at least one topographic image based on the distances between the pairs of points it has determined in the preceding steps. Thus, the processor establishes the discrepancy measure as a weighted measurement value of the set of distances between the pairs of points; that is, between the point in the region of the topographic image and the point in the at least one portion of the patient's anatomy (segmented virtual model).

[0067] In one embodiment, the weighted measurement value is the mean square value of the set of distances between the pairs of points.

[0068] In this embodiment, a particular weighted measurement value is defined, without prejudice to other values ​​having a place in the context of the invention.

[0069] In one embodiment, the topographic imaging module acquires a plurality of topographic images over time, and the tracking module dynamically tracks over time the three-dimensional pose of the at least a portion of the anatomy of a patient, and the discrepancy measurement is performed by comparing: the time evolution of the three-dimensional pose of the region of the topographic images and the time evolution of the three-dimensional pose of the at least a portion of the anatomy of the patient.

[0070] In this embodiment, the topographic imaging module acquires a plurality of topographic images, each at a different time, and sends them to the processor, which establishes the three-dimensional pose of a region in each of them. This region must be the same in the image sequence, whether it is the entire topographic image itself or a portion or part of it.

[0071] On the other hand, the tracking module acquires the three-dimensional pose of at least one part of the patient's anatomy at a plurality of time points and sends said poses to the processor. Under these conditions, the processor calculates the discrepancy by analyzing how the poses of the region of the topographic images vary over time in relation to how the poses of at least one part of the patient's anatomy vary over time. Thus, the processor:

[0072] - calculates the transformation matrix between the three-dimensional pose of the region of at least one topographic image acquired by the topographic image module at a first time instant and the three-dimensional pose of the region of the at least one topographic image acquired by the topographic image module at a second time instant;

[0073] - calculates the transformation matrix between the three-dimensional pose of at least one part of the patient's anatomy acquired by the tracking module at the first time instant and the three-dimensional pose of at least one part of the patient's anatomy acquired by the tracking module at the second time instant; and

[0074] - calculates the discrepancy as the difference between both transformation matrices, differentiating them by measuring them according to a pre-established norm, preferably the Euclidean norm of the vector line between the translation values ​​of the transformation matrices.

[0075] Throughout the document, a transformation matrix between two poses will be understood as a matrix that completely quantifies the angular and translational correspondence between two poses.

[0076] In one embodiment, the acquisition of the three-dimensional pose of at least a portion of the anatomy of a patient by the tracking module comprises carrying out by the processor:

[0077] - either commanding the acquisition of position measurements of a plurality of points on the patient's surface, or commanding the acquisition of position measurements of fiducial markers previously fixed to a part of the patient's anatomy;

[0078] - determining the pose of the virtual model representing at least part of the patient's anatomy and the fiducial points or markers where the measurement is acquired, the determination calculated from the previous measurements on the fiducial points or markers represented in the virtual model; and

[0079] - determining the three-dimensional pose of at least a portion of the patient's anatomy from the pose of the virtual model. Throughout this document, a virtual model of the patient will be understood to be a computational virtual representation of a portion of the patient's anatomy generated from an image of said anatomy. In one embodiment, the virtual model represents the segmentation of one or more tissues of the patient's anatomy, for example a bone, with respect to the rest of the surrounding tissues.

[0080] In one embodiment, if the tracking module comprises sets of fiducial markers:

[0081] - First, it displays the individual markers and acquires measurements of their position.

[0082] - The tracking module then determines the pose of a virtual model that represents at least part of the patient's anatomy and the sets of fiducial markers from which the pose measurement is acquired. This pose of the virtual model is determined using the previous measurements, triangulating the marker positions, and comparing those positions with known geometries declared in the tracking module. In this case, the virtual model is considered to comprise the markers as part of a body that kinetically behaves like a rigid solid.

[0083] - Finally, the three-dimensional pose of at least a part of the patient's anatomy is determined from the pose of the virtual model containing said part of the patient's anatomy.

[0084] In one embodiment, if the tracking module is based on an object tracking algorithm, for example ORB-SLAM:

[0085] - First, it acquires position measurements of points on the patient's surface. This patient's surface is represented by several images acquired by one or more cameras in the tracking module.

[0086] - The tracking module then determines the pose of a virtual model generated from the images acquired by the tracking module's camera(s). The virtual model represents at least part of the patient's anatomy and the points where the measurements are acquired. This pose of the virtual model is determined from variations in the position measurements of points located in several consecutive images.

[0087] - Finally, the three-dimensional pose of at least a part of the patient's anatomy is determined from the pose of the virtual model containing said part of the patient's anatomy.

[0088] In one embodiment, the link that relates the first reference system and the second reference system is established either by a mechanical connection relationship with the relative poses between the first reference system and the second reference system pre-established, or by a measurement of the pose of the topographic image module, integral with the second reference system, with respect to the first reference system.

[0089] The link between the first and second reference systems, which are connected to the tracking module and the topographic image module respectively, can be determined in several ways.

[0090] When the tracking module and the topographic imaging module are mechanically connected, the connection relationship is known at the factory by the processor, so the connection is established directly. If this relationship becomes unbalanced, it is understood that it can be recalibrated.

[0091] When the tracking module and the topographic image module are not mechanically connected, the processor needs to know the pose of the topographic image module, attached to the second reference system, with respect to the first reference system.

[0092] To achieve this, a calibration process can be carried out prior to using the navigation system, which allows the poses of the modules to be determined and thus establishes the relationship between the pose of the topographic image module with respect to the first reference system.

[0093] Alternatively, the tracking module itself may measure the pose of the topographic image module using a set of fiducial markers attached to the topographic image module, such that the processor may establish the relationship between the pose of the topographic image module and the first reference system.

[0094] In one embodiment, the processor comprises: - a first central processing unit;

[0095] - a second processing unit associated with the tracking module to command the acquisition of measurements and to determine the three-dimensional pose of at least one part of the anatomy of a patient from the acquired measurements and, configured to transmit the three-dimensional pose to the first central processing unit;

[0096] - a third processing unit associated with the topographic image module to command the acquisition of measurements and configured to transmit the at least one topographic image to the first central processing unit.

[0097] In this embodiment, the navigation system comprises a distributed processor with three units: a first central unit, a second processing unit associated with the tracking module and a third central unit associated with the topographic image module.

[0098] The second processing unit controls the acquisition of measurements (on fiducial markers or on images acquired by the tracking module's cameras, to locate reference points) and determines the three-dimensional pose of at least a portion of a patient's anatomy from the acquired measurements. Furthermore, said second processing unit transmits the three-dimensional pose to the first central processing unit.

[0099] The third processing unit controls the acquisition of measurements of the patient's anatomy to generate the topographic image. This third processing unit sends the topographic image to the first central processing unit, which determines the three-dimensional pose of a region of said topographic image relative to the second reference system.

[0100] The first processing unit is configured to:

[0101] • establish the link between the first reference system and the second reference system;

[0102] • establish the base reference system as well as its link to the first and second reference systems;

[0103] • receive, from the second processing unit, the three-dimensional pose of at least a part of the patient's anatomy;

[0104] • receiving, from the third processing unit, the at least one topographic image and establishing the three-dimensional pose of a region of the at least one topographic image with respect to the second reference system;

[0105] • referencing the three-dimensional pose of the topographic image region and the three-dimensional pose of at least a part of the patient's anatomy with respect to the base reference system;

[0106] • calculating a discrepancy measure between the three-dimensional pose of at least a portion of the patient's anatomy and the three-dimensional pose of the region of the at least one topographic image; and

[0107] • issue a tracking failure alarm if the discrepancy measurement exceeds a predetermined value.

[0108] In one embodiment, the topographic imaging module is further configured to perform automatic tissue distinction in the at least one topographic image of at least a portion of the patient's anatomy, preferably based on tissue roughness, signal intensity at the tissue surface, or polarimetric contrast in the tissues.

[0109] In this embodiment, the topographic imaging module also automatically distinguishes biological tissues present in the surgical field. This feature is of great interest in the clinical setting, as it allows the detection of critical tissue structures such as nerves or blood vessels that should be avoided during surgery.

[0110] Additionally, the topographic imaging module is capable of discerning structures with complex configurations, such as the distribution of different types of tissue. This more complex structure also allows for the establishment of comparison and matching conditions between images to achieve more precise alignments between them. For example, the topographic imaging module can discern between soft tissue and bone, such that the processor identifies a region in the topographic image identified as bone. This allows the bone poses to be compared when the discrepancy between the pose of this region and the pose of at least part of the patient's anatomy is evaluated.

[0111] In one embodiment, the system further comprises:

[0112] - an actuation module configured to receive at least one surgical tool, and / or

[0113] - a laser cutting module configured to perform a cut in biological tissue in a cut region of the patient's anatomy, wherein the processor is in communication with the laser cutting module and is additionally configured to deactivate the laser cutting module in the event of issuing a tracking failure alarm. The navigation system according to this embodiment may comprise an actuation module configured to receive surgical tools or surgical instruments to be manipulated by surgeons, such that said module acts as a support for such tools and instruments, thereby increasing the precision of their use.

[0114] Additionally or alternatively, the navigation system of this embodiment comprises a laser cutting module for making cuts in the patient's anatomical tissues. Laser cutting has the advantage of high precision and, furthermore, of allowing cuts with arbitrary geometries, of being vibration-free, and of improving tissue recovery. Furthermore, this laser cutting module is advantageously very safe, as it communicates with the processor, such that the processor deactivates the laser if it has issued a tracking failure alarm.Thus, in a case where the navigation system is incorrectly displaying a part of the patient's anatomy that is suitable for laser cutting, but the processor has instead detected that there is a tracking error (the part of the anatomy being displayed in a pose at a given time is not the actual anatomy located in that pose at that time), the laser cutting module would be prevented from making a cut in an unwanted tissue, such as a nerve or a blood vessel.

[0115] In one embodiment, the system additionally comprises a surgical head, and wherein

[0116] - the topographic imaging module is located in the surgical head, and / or

[0117] - the laser cutting module is located in the surgical head, and / or

[0118] - the actuation module is located in the surgical head.

[0119] Advantageously, by having the surgical head, the topographic imaging module, and the module(s) that work on the patient's tissues to perform the surgery (laser cutting module and / or actuation module) in the same physical space, the surgical workflow is simplified. Furthermore, the topographic imaging module will always display the part of the patient's anatomy on which the surgery is being performed, allowing the information to be compared with the information from the monitoring module to always be up-to-date.

[0120] In one embodiment, the topographic imaging module is further configured to automatically adjust its scanning area so that the at least one topographic image it acquires comprises the slice region and the perimeter of the slice region.

[0121] The slice region is a region of the patient's anatomy on which surgeons plan to make at least one slice during the course of the surgery. This region is defined by the surgeon intraoperatively or preoperatively from the virtual model of the patient, for example, by selecting the region from images taken by the topographic imaging module, the tracking module, or external imaging equipment (e.g., a CT scanner).

[0122] The slice region is stored by the processor and sent to the topographic imaging module so it can automatically adjust its scanning area. Thus, the navigation system advantageously ensures that the topographic images show at least the slice region on which the surgery is being performed.

[0123] In an embodiment where the processor is a distributed processor, the cut region is stored by the first central processing unit and sent by the first central processing unit to the third processing unit associated with the topographic imaging module, which is configured to automatically adjust the scanning area of ​​the topographic imaging module.

[0124] In one embodiment, the system additionally comprises a robotic arm, wherein the surgical head is located on said robotic arm.

[0125] The navigation system according to this embodiment is a robotic system comprising a robotic arm into which the surgical head is integrated. This type of system allows surgical actions, such as laser cutting or implant guidance, to be performed with great precision; thus, a small deviation in the position of the patient's anatomy shown to the surgeon could cause serious harm to the patient (for example, by the unwanted cutting of a small nerve shown to the surgeon in the wrong position). Thus, the navigation system of the invention represents a great advantage for this particular type of robotic surgery, as it is capable of detecting very subtle posed inaccuracies that would otherwise go completely unnoticed.

[0126] In one embodiment, the robotic arm comprises at least one sensor configured to estimate the three-dimensional pose of the surgical head relative to the base reference system. - 7.1 -

[0127] By having a robotic arm, in this embodiment it is possible to have additional information about the three-dimensional pose of the surgical head (in which the topographic imaging module could be integrated) thanks to the presence of a sensor in the robotic arm itself.

[0128] In one embodiment, a specific way of measuring the pose of the surgical head consists of measuring, using the robot's encoders, the position of its end-effector, knowing the kinematic chain that configures the robotic arm.

[0129] In one embodiment, the tracking module is further configured to acquire the three-dimensional pose of the topographic imaging module, and wherein the processor is further configured to: receive, from the robotic arm, the three-dimensional pose of the surgical head; receive, from the tracking module, the three-dimensional pose of the topographic imaging module; calculate a discrepancy measure between the three-dimensional pose of the tracking module and the three-dimensional pose of the surgical head; and issue a tracking failure alarm if the discrepancy measure exceeds a predetermined value.

[0130] This embodiment provides an additional option for detecting tracking errors using information provided by the robotic arm sensor. This sensor is configured to estimate the three-dimensional pose of the surgical head, into which the topographic imaging module would be integrated.

[0131] On the other hand, the tracking module is configured to acquire the three-dimensional pose of the topographic image module; for example, by attaching fiducial markers to said topographic image module.

[0132] In this way, the processor receives the three-dimensional pose of the surgical head and the three-dimensional pose of the topographic imaging module to calculate a discrepancy measure between the poses such that, if the discrepancy exceeds a predetermined value, the processor issues a tracking failure alarm.

[0133] In one embodiment, the discrepancy measure is calculated as: - the difference between the three-dimensional pose of the topographic imaging module at a given time instant and the pose it would have under the condition of exhibiting the smallest distance from the three-dimensional pose of the surgical head at that time instant; or

[0134] - the difference between the three-dimensional pose of the surgical head at a given time instant and the pose it would have under the condition of exhibiting the smallest distance from the three-dimensional pose of the topographic imaging module at that time instant; where the distance is measured according to a pre-established standard, preferably the Euclidean standard.

[0135] In one embodiment, the sensor of the robotic arm dynamically tracks the three-dimensional pose of the surgical head over time, and the tracking module dynamically tracks the three-dimensional pose of the topographic imaging module over time, and the discrepancy measurement is performed by comparing: the time evolution of the three-dimensional pose of the surgical head and the time evolution of the three-dimensional pose of the topographic imaging module.

[0136] In this implementation, the comparison of temporal evolutions requires, on the part of the processor:

[0137] - calculating the distance between the three-dimensional pose of the topographic image module at a first time instant and the three-dimensional pose of the topographic image module at a second time instant;

[0138] - calculate the distance between the three-dimensional pose of the surgical head at the first time instant and the three-dimensional pose of the surgical head at the second time instant;

[0139] - calculate the discrepancy as the difference between both distances, the distances being measured according to a pre-established standard, preferably the Euclidean standard.

[0140] In one embodiment, the topographic image module is:

[0141] - an optical coherence tomography (OCT) system; or

[0142] - a PS-OCT polarization-sensitive optical coherence tomography system; or - a stereo pair system; or

[0143] - a structured light system; or

[0144] - an optoacoustic tomography system; or

[0145] - a combination of the above.

[0146] In this embodiment, different types of topographic image modules are defined, without prejudice to the fact that other types of topographic image modules have a place in the context of the present invention.

[0147] In one embodiment, the tracking module is:

[0148] - an optical tracking system with fiducial markers; or

[0149] - a system for tracking points in an image sequence, wherein the tracking of points in the image sequence is performed by a point tracking algorithm executed in the second processing unit, preferably an ORB-SLAM type algorithm.

[0150] In this embodiment, different types of tracking modules are defined, without prejudice to the fact that other types of tracking modules have a place in the context of the present invention.

[0151] All features and / or method steps described herein (including the claims, description and drawings) may be combined in any combination, except for combinations of such features that are mutually exclusive.

[0152] DESCRIPTION OF THE DRAWINGS

[0153] These and other features and advantages of the invention will become more clearly apparent from the following detailed description of a preferred embodiment, given solely as an illustrative and non-limiting example, with reference to the accompanying figures.

[0154] Figures la-lb These figures show a surgical navigation system according to two different embodiments of the invention.

[0155] Figures lc-lf These figures show different ways of calculating three-dimensional pose discrepancy for the surgical navigation systems shown in Figures la-lb. Figures 2a -2b These figures show the surgical navigation system according to two different embodiments of the invention, where the tracking module is based on an image point tracking algorithm (Figure 2a) or on fiducial markers (Figure 2b).

[0156] Figure 3 This figure shows the surgical navigation system according to an embodiment of the invention in which the processor is a distributed processor.

[0157] Figure 4 This figure shows the surgical navigation system according to an embodiment of the invention in which the system comprises a surgical head.

[0158] Figure 5 This figure shows the surgical navigation system according to an embodiment of the invention in which the system comprises a robotic arm.

[0159] DETAILED EXPLANATION OF THE INVENTION

[0160] The Figure shows the surgical navigation system (1) of the invention according to an exemplary embodiment. The system (1) comprises a processor (2); a tracking module (3) configured to acquire the three-dimensional pose of at least a part of the anatomy of a patient (4) with respect to a first reference system (SR1); and a topographical imaging module (5) configured to acquire at least one topographical image of at least a part of the anatomy of the patient, the at least one topographical image (6) being referenced with respect to a second reference system (SR2).

[0161] The processor (2) is in communication with the tracking module (3) and with the topographic image module (5).

[0162] This processor (2) is configured to either establish a link that relates the first reference system (SR1) and the second reference system (SR2); or establish a base reference system (SR) as well as a first link that relates the base reference system (SR) and the first reference system (SR1) and, a second link that relates the base reference system (SR) and the second reference system (SR2).

[0163] In one embodiment, the link relating the first reference system (SR1) and the second reference system (SR2) is established by a mechanical joining relationship with the relative poses between the first reference system (SR1) and the second reference system (SR2) pre-established.

[0164] In another embodiment, the link that relates the first reference system (SR1) and the second reference system (SR2) is established by a measurement of the pose of the topographic image module (5), integral with the second reference system (SR2), with respect to the first reference system (SR1).

[0165] The processor (2) is additionally configured to carry out in any order: a) receive the at least one topographic image (6) from the topographic image module (5) and establish the three-dimensional pose of a region (6.1) of the at least one topographic image (6) with respect to the second reference system (SR2), and b) receive, from the tracking module (3), the three-dimensional pose of at least a part of the patient's anatomy (4).

[0166] The processor (2) is additionally configured to reference the three-dimensional pose of the region (6.1) of the at least one topographic image (6) and the three-dimensional pose of the at least one part of the patient's anatomy (4) with respect to the base reference system (SR).

[0167] Finally, the processor (2) is further configured to: calculate a discrepancy measure between the three-dimensional pose of at least one part of the patient's anatomy (4) acquired by the tracking module (3) and the three-dimensional pose of the region (6.1) of the at least one topographic image (6); and issue a tracking failure alarm if the discrepancy measure exceeds a predetermined value. - TI -

[0168] In one embodiment, the topographic imaging module (5) is further configured to perform automatic tissue distinction in the at least one topographic image (6) of at least a portion of the patient's anatomy, preferably based on tissue roughness, signal intensity at the surface of the tissues, or polarimetric contrast in the tissues.

[0169] In one embodiment, the topographic image module (5) is:

[0170] - an optical coherence tomography (OCT) system; or

[0171] - a polarization-sensitive optical coherence tomography (PS-OCT) system; or

[0172] - a stereo pair system; or

[0173] - a structured light system; or

[0174] - an optoacoustic tomography system; or,

[0175] - a combination of the above.

[0176] In the embodiment of Figure Ib, in addition to the elements shown in Figure la, at least a part of the patient's anatomy (4) is represented by a virtual model (9) generated by the processor (2) from a preoperative or intraoperative image, where the preoperative or intraoperative image is acquired by the topographic image module (5), by the tracking module (3) or by imaging equipment external to the system.

[0177] In one embodiment, the at least one part of the patient's anatomy (4) is a region that corresponds to the best approximation of the region (6.1) of the at least one topographic image (6) in the patient's anatomy, determined as a region that verifies to belong to the surface of the patient's anatomy and have the minimum distance to the region (6.1) of the topographic image (6), the distance measured according to a pre-established norm, preferably the Euclidean norm.

[0178] For any of the embodiments shown in Figures 1a and 1b, the discrepancy measure can be calculated in different ways.

[0179] In the realization of Figure le, the discrepancy measure is calculated as the difference between the three-dimensional pose of at least one part of the patient's anatomy (4) at a given time instant ti and the pose that it would have under the condition of showing the smallest distance d m n with the three-dimensional pose of the region (6.1) of the at least one topographic image (6) at said time instant ti.

[0180] In this embodiment, the processor (2) calculates the discrepancy based on the transformation T that the three-dimensional pose of at least one part of the patient's anatomy (4) would have to undergo so that it would be as similar as possible to the three-dimensional pose of the region (6.1) of the at least one topographic image (6), the poses having been acquired or estimated at the same time instant ti.

[0181] Alternatively, as shown in Figure Id, the discrepancy measure is calculated as the difference between the three-dimensional pose of the region (6.1) of the at least one topographic image (6) at a given time instant ti and the pose it would have under the condition of showing the smallest distance d m ¡ n with the three-dimensional pose of at least a part of the patient's anatomy (4) at said time instant ti.

[0182] In this case, the processor (2) calculates the discrepancy based on the transformation T that the three-dimensional pose of the region (6.1) of the at least one topographic image (6) would have to undergo so that it would be as similar as possible to the three-dimensional pose of at least one part of the patient's anatomy (4), the poses having been acquired or estimated at the same time instant ti.

[0183] For all cases, the distance is measured according to a pre-established standard, preferably the Euclidean standard.

[0184] Figure 1: Discrepancy measure by calculating distances between points on surface regions positioned according to the estimated poses.

[0185] In the embodiment of Figure 1, at least a part of the patient's anatomy (4) is a region resulting from segmenting the virtual model (9), preferably determining a border surface between different tissues such as soft tissue and bone.

[0186] In this case, the processor (2) calculates the discrepancy measure according to the following steps: a) Determining a set of points {pi, P2, Pa} in the region (6.1) of the at least one topographic image (6). In one embodiment, the set of points comprises all the points of the region (6.1) of the at least one topographic image (6). In another embodiment, the set of points comprises a selection of representative points of the region (6.1) of the at least one topographic image (6). b) For each point of the previous step, determining the point {p'i, p'2, p'a} of the at least one part of the patient's anatomy (4) that is verified to have the smallest distance with said point of the region (6.1) of the at least one topographic image (6).That is, for each point in the set {pi, p2, pa}, the processor calculates the distances between said point {pi, p2, pa} and the points of at least one part of the patient's anatomy (4) to identify those that have the shortest distance to the points in the set {pi, p2, pa}. In this example, the points of the part of the patient's anatomy (4) identified as those with the shortest distance from {pi, p2, pa} are, respectively, {p'i, p'2, p'a}.

[0187] The distance between points is measured according to a pre-established standard, preferably the Euclidean standard. c) Establishing the measure of the discrepancy between the three-dimensional pose of at least one part of the patient's anatomy (4) and the three-dimensional pose of the region (6.1) of the at least one topographic image (6) as a weighted measurement value of the set of distances between the pairs of points {pi-p'i, p2-p'2, pa- p'a} determined in the previous stage; that is, the point of the region (6.1) of the topographic image (6) and the point of at least one part of the patient's anatomy (4) or segmented virtual model (9).

[0188] In one embodiment, the weighted measure value is the mean square value of the set of distances between the pairs of points. In this example of Figure 11, the measure value d RM s weighted is calculated:

[0189] Where N is the number of distances between points calculated (in this case three) and d¡ represents each of the distances between pairs of points {pi-p'i, p2-p'2, pa- p'a}. Figure 11: Discrepancy measure comparing evolutions of estimated poses.

[0190] In the embodiment of Figure lf, the topographic image module (5) acquires a plurality of topographic images (6) over time, each at a different time instant, and sends them to the processor (2). The processor (2), in turn, establishes the three-dimensional pose of a region (6.1) in each of them (6). This region (6.1) must be the same in the sequence of images, whether it is the topographic image (6) itself in its entirety or a part or portion thereof.

[0191] In this embodiment, the tracking module (3) further establishes a dynamic tracking over time of the three-dimensional pose of at least one part of the patient's anatomy (4). To do this, the processor (2) acquires the three-dimensional pose of at least one part of the patient's anatomy (4) at a plurality of time instants and sends said poses to the processor (2).

[0192] For simplicity, this example shows only the poses of the regions (6.1) of the topographic images (6) acquired at two different time instants ti and t2. Similarly, this example shows the poses of at least one part of the patient's anatomy (4) between said two time instants ti and t2.

[0193] Taking into account this temporal information, the processor (2) calculates the discrepancy measure by comparing the temporal evolution of the pose of the region (6.1) of the topographic images (6) and the temporal evolution of the pose of at least a part of the patient's anatomy (4). In particular, the processor (2):

[0194] - calculates the transformation matrix Mi between the three-dimensional pose of the region (6.1) of at least one topographic image (6) acquired by the topographic image module (5) at a first time instant ti and the three-dimensional pose of the region (6.1) of the at least one topographic image (6) acquired by the topographic image module (5) at a second time instant t2;

[0195] - calculates the transformation matrix M2 between the three-dimensional pose of at least one part of the patient's anatomy (4) acquired by the tracking module (3) at the first time instant ti and the three-dimensional pose of at least one part of the patient's anatomy (4) acquired by the tracking module (3) at the second time instant t2; and

[0196] - calculates the discrepancy as the difference between both transformation matrices Mi and M2, the difference being measured according to a pre-established norm, preferably the Euclidean norm of the vector line between the translation values ​​of the transformation matrices.

[0197] Two different embodiments of the navigation system of the invention are shown in Figures 2a and 2b, each comprising a different type of tracking module (3).

[0198] Figure 2a shows an embodiment of the system (1) of the invention in which the tracking module (3) is based on a point tracking algorithm or object tracking in images.

[0199] This tracking module (3) comprises one or more cameras (not shown in Figure 2a) that acquire a plurality of images of the surgical field and a processing unit, which is part of the processor (2) itself, capable of executing an object tracking algorithm, preferably the ORB-SLAM algorithm, to acquire the three-dimensional pose of at least a part of the patient's anatomy (4).

[0200] In particular, the acquisition of the three-dimensional pose of at least one part of the anatomy of a patient (4) comprises carrying out by means of the processor (2): commanding, to the tracking module (3), the acquisition of position measurements of a plurality of points (7) of the surface of the patient; determining the pose of the virtual model (9) that represents the at least one part of the anatomy of the patient (4) and the points (7) where the measurement is acquired, the determination calculated from the previous measurements on the points (7) represented in the virtual model (9); and determining the three-dimensional pose of the at least one part of the anatomy of the patient (4) from the pose of the virtual model (9).

[0201] The pose of the virtual model (9) is determined by the object tracking algorithm, which selects a plurality of reference points in a first image of a portion of the patient's anatomy (4) taken by one or more cameras and subsequently locates all or part of said reference points in images taken by one or more cameras at subsequent time points. Thus, the tracking module tracks the virtual model (9) - and therefore the portion of the patient's anatomy (4) - by measuring variations in the position of the reference points (generally translations and rotations) located in consecutive images, and estimates the three-dimensional pose of the virtual model (9) and of said anatomy (4) over time.

[0202] Figure 2b shows an embodiment of the system (1) of the invention in which the tracking module (3) is based on the display of fiduciary markers (8).

[0203] This tracking module (3) comprises an infrared stereo camera (not shown in Figure 2b) and a plurality of fiducial markers (8), preferably in the form of reflective spheres and / or LEDs (Light Emitting Diodes), arranged in marker arrays of known geometries. These fiducial marker arrays (8) are fixed to a part of the patient's anatomy in such a way that they are visible to the tracking module (3), which is configured to determine their positions by triangulation, compare said positions with geometries known to the tracking module (3), and determine the three-dimensional pose of the part of the patient's anatomy (4) to which they have been fixed.

[0204] In this embodiment, the acquisition of the three-dimensional pose of at least a part of the anatomy of a patient (4) by the tracking module (3) comprises carrying out by means of the processor (2): commanding the acquisition of position measurements of fiduciary markers (8) previously fixed to a part of the anatomy of the patient; determining the pose of the virtual model (9) that represents the at least a part of the anatomy of the patient (4) and the fiduciary markers (8) where the measurement is acquired, the determination calculated from the previous measurements on the fiduciary markers (8) represented in the virtual model (9); and determining the three-dimensional pose of the at least a part of the anatomy of the patient (4) from the pose of the virtual model (9).

[0205] In one embodiment, further sets of fiducial markers (8) are attached to the topographic imaging module (5) such that they are visible to the tracking module (3) which is configured to acquire the three-dimensional pose of the topographic imaging module (5) over time. Figure 3 shows an embodiment of the navigation system (1) of the invention wherein the processor (2) is a distributed processor. In this case, the processor (2) comprises:

[0206] - a first central processing unit (2.1);

[0207] - a second processing unit (2.2) associated with the tracking module (3) to command the acquisition of measurements and to determine the three-dimensional pose of at least one part of the anatomy of a patient (4) from the acquired measurements and, configured to transmit the three-dimensional pose to the first central processing unit (2.1); and

[0208] - a third processing unit (2.3) associated with the topographic image module (5) to command the acquisition of measurements and configured to transmit the at least one topographic image (6) to the first central processing unit (2.1).

[0209] The first processing unit (2.1) is configured to:

[0210] • establish the link between the first reference system (SR1) and the second reference system (SR2);

[0211] • establish the base reference system (SR) as well as its link to the first and second reference systems (SRI, SR2);

[0212] • receive, from the second processing unit (2.2), the three-dimensional pose of at least one part of the patient's anatomy (4);

[0213] • receive, from the third processing unit (2.3), the at least one topographic image (6) and establish the three-dimensional pose of a region (6.1) of the at least one topographic image (6) with respect to the second reference system (SR2);

[0214] • referencing the three-dimensional pose of the region (6.1) of the topographic image (6) and the three-dimensional pose of at least a part of the patient's anatomy (4) with respect to the base reference system (SR);

[0215] • calculating a discrepancy measure between the three-dimensional pose of at least a portion of the patient's anatomy (4) and the three-dimensional pose of the region (6.1) of the at least one topographic image (6); and

[0216] • issue a tracking failure alarm if the discrepancy measurement exceeds a predetermined value.

[0217] Figure 4 shows the navigation system (1) according to an embodiment of the invention in which the system (1) comprises a surgical head (11).

[0218] The system (1) shown in Figure 4 additionally comprises: - an actuation module (10) configured to receive at least one surgical tool, and / or

[0219] - a laser cutting module (10') configured to perform a cut in the biological tissues of a cutting region of the patient's anatomy, wherein the processor (2) is in communication with the laser cutting module (10') and is additionally configured to deactivate the laser cutting module (10') in case of issuing a tracking failure alarm.

[0220] In the embodiment shown in Figure 4, the topographic imaging module (5) as well as the actuation module (10) and / or the laser cutting module (10') are integrated into the surgical head (11).

[0221] In other embodiments not shown, the navigation system (1) comprises an actuation module (10) and / or a laser cutting module (10'), but these are not integrated into a surgical head (11).

[0222] In an embodiment wherein the system (1) comprises a laser cutting module (10') configured to perform a cut in biological tissues of a cut region of the patient's anatomy, the topographic imaging module (5) is further configured to automatically adjust its scanning area so that the at least one topographic image (6) it acquires comprises the cut region and the perimeter of the cut region. The laser cutting module (10') and the topographic imaging module (5) of this embodiment may or may not be integrated into a surgical head (11).

[0223] A navigation system (1) according to an embodiment is shown in Figure 5, wherein the system (1) additionally comprises a robotic arm (12). The surgical head (11) described in Figure 4 is located on said robotic arm (12) in this embodiment.

[0224] In one embodiment, the robotic arm (12) comprises at least one sensor configured to estimate the three-dimensional pose of the surgical head (11) with respect to the base reference system (SR).

[0225] In one embodiment, the tracking module (3) is further configured to acquire the three-dimensional pose of the topographic imaging module (5), and the processor (2) is further configured to: receive, from the robotic arm (12), the three-dimensional pose of the surgical head (11); receive, from the tracking module (3), the three-dimensional pose of the topographic imaging module (5); calculate a discrepancy measure between the three-dimensional pose of the tracking module (3) and the three-dimensional pose of the surgical head (11); and issue a tracking failure alarm if the discrepancy measure exceeds a predetermined value.

[0226] Below are different specific non-limiting examples of the invention:

[0227] Navigation system (1) with tracking module (3) based on fiduciary markers (8) and with PS-OCT system (5).

[0228] In system (1) of this example:

[0229] - the tracking module (3) measures the three-dimensional pose of at least a part of the patient's anatomy (4) from the visualization of the set of fiducial markers (8) fixed to said patient's anatomy, and

[0230] - the topographic image module (5) is a PS-OCT system that takes measurements of a volume of a part of the patient's anatomy and the processor (2) establishes the topographic image (6) as the external surface delimiting the part of the patient's anatomy included within the measured volume, determined by segmentation, generating for this purpose a numerical model that represents said surface.

[0231] In this example, the tracking module (3) estimates the pose of the topographic image module (5) by viewing another set of fiduciary markers (8) attached to said module (5). In this way, the processor (2) is able to establish the link that relates the first reference system (SR1), integral with the tracking module (3), and the second reference system (SR2), integral with the topographic image module (5).

[0232] In this example, the at least part of the patient's anatomy (4) is a region resulting from segmenting the virtual model (9) generated by the processor (2) from a preoperative computed tomography image. In this example, the virtual model (9) comprises a bone of the patient segmented with respect to the surrounding soft tissues. The virtual model (9) also presents the fiducial markers (8). The processor (2) of this system (1) knows the three-dimensional pose of the virtual model (9) - of the part of the patient's anatomy (4) - determined by the tracking module (3) and establishes the three-dimensional pose of a region (6.1) of the topographic image (6) acquired by the PS-OCT system (5).

[0233] The processor (2) transforms the pose of the region (6.1) of the topographic image (6) according to the pose of the virtual model (9) segmented by means of a registration procedure, widely known in the field of surgical navigation.

[0234] At this point, the processor calculates the discrepancy measure by comparing pairs of points in the region (6.1) of the topographic image (6) and the segmented virtual model (9). To do this:

[0235] - determines a set of points in the region (6.1) of the at least one topographic image (6),

[0236] - for each point of the previous step, determine the point of at least one part of the patient's anatomy (4) that is verified to have the smallest distance with said point of the region (6.1) of the at least one topographic image (6), and

[0237] - establishes the measure of discrepancy as the mean square value of the sum of the distances between the pairs of points determined in the previous stage.

[0238] Additionally, in this example, the system (1) comprises a laser cutting module (10') located in a surgical head (11) next to the topographic imaging module (5). Said surgical head (11), in turn, is located in a robotic arm (12).

[0239] The topographic image module (5) is additionally configured to automatically adjust its scanning area so that the at least one topographic image (6) it acquires comprises a cutting region in which the cutting module (10') performs the cutting of biological tissues, and the perimeter of the cutting region.

[0240] The robotic arm (12) comprises a sensor configured to estimate the three-dimensional pose of the surgical head (11) with respect to the base reference system (SR) established by the processor (2).

[0241] Finally, the processor (2) is additionally configured to: receive, from the robotic arm (12), the three-dimensional pose of the surgical head (11); receive, from the tracking module (3), the three-dimensional pose of the topographic imaging module (5); calculate a discrepancy measure between the three-dimensional pose of the tracking module (3) and the three-dimensional pose of the surgical head (11); and issue a tracking failure alarm if the discrepancy measure exceeds a predetermined value.

[0242] Navigation system (1) with tracking module (3) based on fiducial markers (8) and with stereo pair system (5).

[0243] In system (1) of this example:

[0244] - the tracking module (3) measures the three-dimensional pose of at least a part of the patient's anatomy (4) from the visualization of the set of fiducial markers (8) fixed to said patient's anatomy, and

[0245] - the topographic image module (5) is a stereo pair system that takes measurements of a surface of a part of the patient's anatomy and the processor (2) establishes the topographic image (6) as said measured surface, generating for this purpose a numerical model that represents the surface.

[0246] In this example, the tracking module (3) estimates the pose of the topographic image module (5) by viewing another set of fiduciary markers (8) attached to said module (5). In this way, the processor (2) is able to establish the link that relates the first reference system (SR1), integral with the tracking module (3), and the second reference system (SR2), integral with the topographic image module (5).

[0247] In this example, at least part of the patient's anatomy (4) is represented by a virtual model (9) generated by the processor (2) from an intraoperative image acquired by the topographic imaging module (5) itself. The virtual model (9) also presents the fiducial markers (8).

[0248] The processor (2) of this system (1) knows the three-dimensional pose of the virtual model (9) - of the part of the patient's anatomy (4) - determined by the tracking module and establishes the three-dimensional pose of a region (6.1) of the topographic image (6) acquired by the stereo pair. At this point, the processor (2) calculates the discrepancy measure as the difference between the three-dimensional pose of the at least one part of the patient's anatomy (4) at a given time instant and the pose that it would have under the condition of showing the smallest distance with the three-dimensional pose of the region (6.1) of the at least one topographic image (6) at said time instant.

[0249] Alternatively, the processor (2) calculates the discrepancy measure as the difference between the three-dimensional pose of the region (6.1) of the at least one topographic image (6) at a given time instant and the pose that it would have under the condition of showing the smallest distance with the three-dimensional pose of the at least one part of the patient's anatomy (4) at said time instant.

[0250] In both cases, the distance is measured according to a pre-established standard, preferably the Euclidean standard.

[0251] Additionally, in this example, the system (1) comprises a laser cutting module (10') located in a surgical head (11) next to the topographic imaging module (5). Said surgical head (11), in turn, is located in a robotic arm (12).

[0252] The topographic image module (5) is additionally configured to automatically adjust its scanning area so that the at least one topographic image (6) it acquires comprises a cutting region in which the cutting module (10') performs the cutting of biological tissues, and the perimeter of the cutting region.

[0253] The robotic arm (12) comprises a sensor configured to estimate the three-dimensional pose of the surgical head (11) with respect to the base reference system (SR) established by the processor (2).

[0254] Finally, the processor (2) is additionally configured to: receive, from the robotic arm (12), the three-dimensional pose of the surgical head (11); receive, from the tracking module (3), the three-dimensional pose of the topographic imaging module (5); calculate a discrepancy measure between the three-dimensional pose of the tracking module (3) and the three-dimensional pose of the surgical head (11); and issue a tracking failure alarm if the discrepancy measure exceeds a predetermined value.

[0255] Navigation system (1) with tracking module (3) based on point tracking algorithm (7), ORB-SLAM, and with OCT system (5).

[0256] In system (1) of this example:

[0257] - the tracking module (3) measures the three-dimensional pose over time of at least a part of the patient's anatomy (4) by executing a point-tracking ORB-SLAM algorithm (7), and

[0258] - the topographic image module (5) is an OCT system that takes measurements of a volume of a part of the patient's anatomy and the processor (2) establishes each topographic image (6) as the surface of the measured volume, generating for this purpose a numerical model that represents said surface.

[0259] In this example, the processor (2) comprises:

[0260] - a first central processing unit (2.1);

[0261] - a second processing unit (2.2) associated with the tracking module (3) for the execution of the ORB-SLAM algorithm; and

[0262] - a third processing unit (2.3) associated with the topographic image module (5).

[0263] In this example, the tracking module (3) and the topographic image module (5) are mechanically connected, so the processor (2) establishes the link that relates the first reference system (SR1), integral with the tracking module (3), and the second reference system (SR2), integral with the topographic image module (5), from this connection.

[0264] In this example, at least part of the patient's anatomy (4) is represented by a virtual model (9) generated by the processor (2) from an intraoperative image acquired by the tracking module (3) itself. The virtual model (9) also presents the reference points (7) found by the ORB-SLAM algorithm.

[0265] The processor (2) of this system (1) knows the three-dimensional pose of the virtual model (9) - of the part of the patient's anatomy (4) - determined by the tracking module (3) and establishes the three-dimensional pose of a region (6.1) of the topographic image (6) acquired by the OCT system. At this point, the processor (2) calculates the discrepancy measure by comparing the temporal evolution of the pose of the region (6.1) of the topographic images (6) and the temporal evolution of the pose of at least one part of the patient's anatomy (4). To do this, the processor (2):

[0266] - calculates the transformation matrix between the three-dimensional pose of the region (6.1) of at least one topographic image (6) acquired by the OCT at a first time instant and the three-dimensional pose of the region (6.1) of the at least one topographic image (6) acquired by the OCT at a second time instant;

[0267] - calculates the transformation matrix between the three-dimensional pose of at least one part of the patient's anatomy (4) acquired by the tracking module (3) at the first time instant and the three-dimensional pose of at least one part of the patient's anatomy (4) acquired by the tracking module (3) at the second time instant; and

[0268] - calculates the discrepancy as the difference between both transformation matrices, differentiating them by measuring them according to a pre-established norm, preferably the Euclidean norm of the vector line between the translation values ​​of the transformation matrices.

[0269] Additionally, in this example the system (1) comprises a laser cutting module (10') located in a surgical head (11) next to the topographic imaging module (5). Said surgical head (11), in turn, is located in a robotic arm (12).

[0270] The topographic image module (5) is additionally configured to automatically adjust its scanning area so that the at least one topographic image (6) it acquires comprises a cutting region in which the cutting module (10') performs the cutting of biological tissues, and the perimeter of the cutting region.

Claims

CLAIMS 1. A surgical navigation system (1) comprising: a processor (2); a tracking module (3) configured to acquire a three-dimensional pose of at least a portion of the anatomy of a patient (4) relative to a first reference system (SR1); a topographical imaging module (5) configured to acquire at least one topographical image (6) of at least a portion of the anatomy of the patient, the at least one topographical image (6) being referenced relative to a second reference system (SR2); the processor (2) is in communication with the tracking module (3) and with the topographical imaging module (5), the processor (2) being configured to establish a link relating the first reference system (SR1) and the second reference system (SR2);or establishing a base reference system (SR) as well as a first link relating the base reference system (SR) and the first reference system (SR1), and a second link relating the base reference system (SR) and the second reference system (SR2); or carrying out in any order a) receiving the at least one topographic image (6) from the topographic image module (5) and establishing the three-dimensional pose of a region (6.1) of the at least one topographic image (6) with respect to the second reference system (SR2), and b) receiving, from the tracking module (3), the three-dimensional pose of at least a part of the patient's anatomy (4); or referencing the three-dimensional pose of the region (6.1) of the at least one topographic image (6) and the three-dimensional pose of the at least one part of the patient's anatomy (4) with respect to the base reference system (SR);and wherein the processor (2) is additionally configured to: o calculate a discrepancy measure between the three-dimensional pose of the at least one part of the patient's anatomy (4) acquired by the tracking module (3) and the three-dimensional pose of the region (6.1) of the at least one topographic image (6); and; or issue a tracking failure alarm if the discrepancy measure exceeds a predetermined value. 2.- System (1) according to the previous claim, wherein at least a part of the patient's anatomy (4) is represented by a virtual model (9) generated by the processor (2) from a preoperative or intraoperative image, where the preoperative or intraoperative image is acquired by the topographic image module (5), by the tracking module (3) or by imaging equipment external to the system. 3.- System (1) according to any of the preceding claims, wherein the at least one part of the anatomy of the patient (4) is a region that corresponds to the best approximation of the region (6.1) of the at least one topographic image (6) in the anatomy of the patient, determined as a region that verifies that it belongs to the surface of the anatomy of the patient and has the minimum distance to the region (6.1) of the topographic image (6), the distance measured according to a pre-established norm, preferably the Euclidean norm. 4.- System (1) according to any of the preceding claims, wherein the discrepancy measure is calculated as: - the difference between the three-dimensional pose of the at least one part of the patient's anatomy (4) at a given time instant and the pose it would have under the condition of showing the smallest distance with the three-dimensional pose of the region (6.1) of the at least one topographic image (6) at said time instant; or - the difference between the three-dimensional pose of the region (6.1) of the at least one topographic image (6) at a given time instant and the pose that it would have under the condition of showing the smallest distance with the three-dimensional pose of the at least one part of the patient's anatomy (4) at said time instant; where the distance is measured according to a pre-established standard, preferably the Euclidean standard. 5.- System according to any of claims 2-4, wherein at least a part of the patient's anatomy is a region resulting from segmenting the virtual model (9), preferably determining a border surface between different tissues, and where the discrepancy measure is calculated according to the following steps: - determining a set of points in the region (6.1) of the at least one topographic image (6), or all the points of the region (6.1) of the at least one image topographic (6) or a selection of representative points of the region (6.1) of the at least one topographic image (6); - for each point of the previous step, determining the point of the at least one part of the patient's anatomy (4) that is verified to have the smallest distance with said point of the region (6.1) of the at least one topographic image (6); - establishing the measure of the discrepancy between the three-dimensional pose of the at least one part of the patient's anatomy (4) and the three-dimensional pose of the region (6.1) of the at least one topographic image (6) as a weighted measurement value of the set of distances between the pairs of points determined in the previous step, the point of the region (6.1) of the topographic image (6) and the point of the at least one part of the patient's anatomy (4). 6.- System according to the previous claim, where the weighted measurement value is the mean square value of the set of distances between the pairs of points. 7.- System (1) according to any of claims 1-3, wherein: - the topographic image module (5) acquires a plurality of topographic images (6) over time and, - the tracking module (3) establishes a dynamic tracking over time of the three-dimensional pose of at least one part of the anatomy of a patient (4) and, where the measurement of the discrepancy is carried out by comparing: the temporal evolution of the three-dimensional pose of the region (6.1) of the topographic images (6) and, the temporal evolution of the three-dimensional pose of at least one part of the anatomy of the patient (4).

8. System (1) according to any of claims 2-7, wherein the acquisition of the three-dimensional pose of at least a part of the anatomy of a patient (4) by the tracking module (3) comprises carrying out by means of the processor (2): either commanding the acquisition of position measurements of a plurality of points (7) of the patient's surface, or commanding the acquisition of position measurements of fiduciary markers (8) previously fixed to a part of the patient's anatomy; determining the pose of the virtual model (9) that represents the at least a part of the patient's anatomy and the points (7) or fiduciary markers (8) where it is acquired the measurement, the determination calculated from the previous measurements on the points (7) or fiducial markers (8) represented in the virtual model (9); and determining the three-dimensional pose of at least a part of the patient's anatomy (4) from the pose of the virtual model (9). 9.- System (1) according to any of the previous claims, wherein the link that relates the first reference system (SR1) and the second reference system (SR2) is established, either by means of a mechanical connection relationship with the relative poses between the first reference system (SR1) and the second reference system (SR2) pre-established, or by means of a measurement of the pose of the topographic image module (5), integral with the second reference system (SR2), with respect to the first reference system (SR1). 10.- System (1) according to any of the preceding claims, wherein the processor (2) comprises: - a first central processing unit (2.1); - a second processing unit (2.2) associated with the tracking module (3) to command the acquisition of measurements and to determine the three-dimensional pose of at least one part of the anatomy of a patient (4) from the acquired measurements and, configured to transmit the three-dimensional pose to the first central processing unit (2.1); - a third processing unit (2.3) associated with the topographic image module (5) to command the acquisition of measurements and configured to transmit the at least one topographic image (6) to the first central processing unit (2.1). 11.- System (1) according to any of the preceding claims, wherein the topographic image module (5) is additionally configured to perform an automatic tissue distinction in the at least one topographic image (6) of at least a part of the patient's anatomy, preferably based on the roughness of the tissues, the intensity of the signal on the surface of the tissues or the polarimetric contrast in the tissues. 12.- System (1) according to any of the preceding claims, which additionally comprises: - an actuation module (10) configured to receive at least one surgical tool, and / or - a laser cutting module (10') configured to perform a cut in the biological tissues of a cut region of the patient's anatomy, wherein the processor (2) is in communication with the laser cutting module (10') and is additionally configured to deactivate the laser cutting module (10') in case of issuing a tracking failure alarm. 13.- System (1) according to the previous claim which additionally comprises a surgical head (11), and wherein - the topographic imaging module (5) is located in the surgical head, and / or - the laser cutting module (10') is located in the surgical head, and / or - the actuation module (10) is located in the surgical head. 14.- System (1) according to any of claims 12-13, wherein the topographic image module (5) is additionally configured to automatically adjust its scanning area so that the at least one topographic image (6) it acquires comprises the cutting region and the perimeter of the cutting region. 15.- System (1) according to any of claims 13-14, which additionally comprises a robotic arm (12), wherein the surgical head (11) is located on said robotic arm (12). 16.- System (1) according to the preceding claim, wherein the robotic arm (12) comprises at least one sensor configured to estimate the three-dimensional pose of the surgical head (11) with respect to the base reference system (SR).

17. System (1) according to the preceding claim, wherein the tracking module (3) is additionally configured to acquire the three-dimensional pose of the topographic imaging module (5) and wherein the processor (2) is additionally configured to: receive, from the robotic arm (12), the three-dimensional pose of the surgical head (11); receive, from the tracking module (3), the three-dimensional pose of the topographic imaging module (5); calculate a discrepancy measure between the three-dimensional pose of the tracking module (3) and the three-dimensional pose of the surgical head (11); and issue a tracking failure alarm if the discrepancy measure exceeds a predetermined value. 18.- System (1) according to any of the preceding claims, wherein the module of topographic image (5) is: - an optical coherence tomography (OCT) system; or - a polarization-sensitive optical coherence tomography (PS-OCT) system; or - a stereo pair system; or - a structured light system; or - an optoacoustic tomography system; or - a combination of the above. 19.- System (1) according to any of the preceding claims and claim 10, wherein the tracking module (3) is: - an optical tracking system with fiducial markers (8); or - a system for tracking points in an image sequence, wherein the tracking of points in the image sequence is performed by a point tracking algorithm executed in the second processing unit (2.1), preferably an ORB-SLAM type algorithm.

Images