Three-dimensional registration method, system and computer equipment for transcranial magnetic stimulation navigation process

Abstract

The present application is applicable to the field of medical technology, and provides a three-dimensional registration method, system and computer equipment for the transcranial magnetic stimulation navigation process, providing a feasible technical solution for the three-dimensional registration process in the transcranial magnetic stimulation navigation technology. The method of the present application mainly includes: determining the first transformation relationship of the probe coordinate system in the camera coordinate system and the second transformation relationship of the head tracking coordinate system where the patient's head is located in the camera coordinate system; when the probe needle tip of the probe coordinate system touches the actual feature point on the patient's head, the coordinates of the actual feature point in the head tracking coordinate system are calculated using the first transformation relationship and the second transformation relationship, and then the virtual feature point corresponding to the actual feature point in the three-dimensional virtual head model is registered to obtain the third transformation relationship between the head tracking coordinate system and the three-dimensional virtual head model coordinate system, and the three-dimensional virtual head model is mapped to the head tracking coordinate system using the third transformation relationship to achieve registration of the patient's head with the three-dimensional virtual head model.

Description

Technical Field

[0001] The present application belongs to the field of medical technology, and in particular relates to a three-dimensional registration method, system and computer equipment for a transcranial magnetic stimulation navigation process. Background Art

[0002] Transcranial magnetic stimulation (TMS) is a non-invasive neuromodulation technology with no clear side effects. Its basic principle is to use a pulsed magnetic field to act on the central nervous system (mainly the cerebral cortex). The induced current generated by the pulsed magnetic field will change the membrane potential of cortical nerve cells, thereby affecting metabolic activity and neural activity in the brain. Currently, the stimulation modes of transcranial magnetic stimulation technology mainly include single pulse, double pulse and repetitive pulse. Single pulse and double pulse stimulation modes are commonly used in routine electrophysiological examinations. Repetitive pulse mode can be used to treat movement disorders, mental illnesses, pathological pain, epilepsy, addiction, and functional recovery after damage to the nervous system.

[0003] During the use of transcranial magnetic stimulation (TMS) technology, 3D registration is a crucial component of TMS navigation. This involves aligning the 3D head model reconstructed from imaging data in the TMS system with the patient's head. Once successfully registered, the 3D head model can accurately guide researchers or doctors during treatment or research.

[0004] However, there is currently no feasible three-dimensional registration method for transcranial magnetic stimulation navigation. Summary of the Invention

[0005] The purpose of this application is to provide a three-dimensional registration method, system and computer equipment for the transcranial magnetic stimulation navigation process, filling the gap in the current three-dimensional registration technology for the transcranial magnetic stimulation navigation process.

[0006] In a first aspect, the present application provides a three-dimensional registration method for a transcranial magnetic stimulation navigation process, comprising:

[0007] Determine the first transformation relationship between the probe coordinate system and the camera coordinate system;

[0008] Determine a second transformation relationship between the head tracking coordinate system represented by the marker points on the patient's head and the camera coordinate system;

[0009] When the probe tip of the probe coordinate system touches X actual feature points on the patient's head respectively, coordinates of the X actual feature points in the camera coordinate system are calculated using the first transformation relationship, where X is a positive integer greater than or equal to 3, and at least three of the X actual feature points are not coplanar or collinear;

[0010] Calculating the coordinates of the X actual feature points in the head tracking coordinate system using the second transformation relationship;

[0011] Picking X virtual feature points corresponding to the X actual feature points in a preset three-dimensional virtual head model, wherein the three-dimensional virtual head model is located in a three-dimensional virtual head model coordinate system;

[0012] respectively registering X virtual feature points corresponding to the X actual feature points to obtain a third transformation relationship between the head tracking coordinate system and the three-dimensional virtual head model coordinate system;

[0013] The three-dimensional virtual head model is mapped to the head tracking coordinate system using the third transformation relationship, so that the patient's head is aligned with the three-dimensional virtual head model.

[0014] Optionally, determining a first transformation relationship of the probe coordinate system in the camera coordinate system includes:

[0015] Acquire a probe image of K markers of the probe in the probe coordinate system, wherein the K markers can reflect the contour shape of the probe, wherein K is a positive integer greater than or equal to 3, and at least three of the K markers are not coplanar and not collinear;

[0016] The first transformation relationship between the probe coordinate system and the camera coordinate system is determined according to the K marking points of the probe image.

[0017] Optionally, K is equal to 4, the contour of the probe is in a Y shape, the four marking points form the Y-shaped contour, and determining the first transformation relationship of the probe coordinate system in the camera coordinate system based on the K marking points of the probe image includes:

[0018] The shape features composed of the K marking points of the probe image are identified and classified by a trained neural network model to obtain the first transformation relationship between the probe coordinate system and the camera coordinate system.

[0019] Optionally, the K marking points include a probe tip of the probe, and after obtaining the first transformation relationship between the probe coordinate system and the camera coordinate system, the method further includes:

[0020] The spatial coordinates of the probe tip in the camera coordinate system are calculated using the first transformation relationship.

[0021] Optionally, determining the second transformation relationship of the head tracking coordinate system represented by the marker points on the patient's head in the camera coordinate system includes:

[0022] Acquiring a head tracking coordinate system image of L markers on the patient's head in the head tracking coordinate system, where the L markers can reflect a spatial shape of the head tracking coordinate system, where L is a positive integer greater than or equal to 3, and at least three of the L markers are not coplanar or collinear;

[0023] A second transformation relationship between the head tracking coordinate system and the camera coordinate system is determined according to the L marking points of the head tracking coordinate system image.

[0024] Optionally, after determining the second transformation relationship between the head tracking coordinate system and the camera coordinate system, the method further includes:

[0025] Acquire marker point images of the L marker points on the patient's head;

[0026] The running trajectory of the head tracking coordinate system reflected in the marker point image is analyzed to obtain motion compensation coordinates of the head tracking coordinate system.

[0027] Optionally, before picking X virtual feature points in the preset three-dimensional virtual head model that respectively correspond to the X actual feature points, the method further includes:

[0028] Acquiring an MRI image of the patient's head using a standard template;

[0029] converting the magnetic resonance image into a bitmap image;

[0030] The three-dimensional virtual head model of the patient's head is constructed using the bitmap image.

[0031] Optionally, after constructing the three-dimensional virtual head model of the patient's head using the bitmap image, the method further includes:

[0032] The three-dimensional virtual head model is registered with the template data of the standard electrode placement method to obtain the 10-20 coordinates on the three-dimensional virtual head model.

[0033] In a second aspect, the present application provides a three-dimensional registration system for a transcranial magnetic stimulation navigation process, comprising:

[0034] A first determining unit, configured to determine a first transformation relationship between the probe coordinate system and the camera coordinate system;

[0035] a second determining unit, configured to determine a second transformation relationship between the head tracking coordinate system represented by the marker points on the patient's head and the camera coordinate system;

[0036] a first calculation unit, configured to calculate, when the probe tip of the probe coordinate system touches X actual feature points on the patient's head respectively, coordinates of the X actual feature points in the camera coordinate system using the first transformation relationship, where X is a positive integer greater than or equal to 3, and at least three of the X actual feature points are not coplanar or collinear;

[0037] A second calculation unit is configured to calculate the coordinates of the X actual feature points in the head tracking coordinate system using the second transformation relationship;

[0038] a picking unit, configured to pick up X virtual feature points corresponding to the X actual feature points in a preset three-dimensional virtual head model, wherein the three-dimensional virtual head model is located in a three-dimensional virtual head model coordinate system;

[0039] a registration unit, configured to respectively register the X virtual feature points corresponding to the X actual feature points, to obtain a third transformation relationship between the head tracking coordinate system and the three-dimensional virtual head model coordinate system;

[0040] A mapping unit is configured to map the three-dimensional virtual head model to the head tracking coordinate system using the third transformation relationship, so that the patient's head is aligned with the three-dimensional virtual head model.

[0041] Optionally, when the first determining unit determines the first transformation relationship of the probe coordinate system in the camera coordinate system, it is specifically used to:

[0042] Acquire a probe image of K markers of the probe in the probe coordinate system, wherein the K markers can reflect the contour shape of the probe, wherein K is a positive integer greater than or equal to 3, and at least three of the K markers are not coplanar and not collinear;

[0043] The first transformation relationship between the probe coordinate system and the camera coordinate system is determined according to the K marking points of the probe image.

[0044] Optionally, K is equal to 4, the contour of the probe is Y-shaped, and the four marking points form the Y-shaped contour. When the first determination unit determines the first transformation relationship of the probe coordinate system in the camera coordinate system according to the K marking points of the probe image, it is specifically used to:

[0045] The shape features composed of the K marking points of the probe image are identified and classified by a trained neural network model to obtain the first transformation relationship between the probe coordinate system and the camera coordinate system.

[0046] Optionally, the K marking points include a probe tip of the probe, and the system further includes:

[0047] A third calculation unit is configured to calculate the spatial coordinates of the probe tip in the camera coordinate system by using the first transformation relationship.

[0048] Optionally, when the second determining unit determines the second transformation relationship of the head tracking coordinate system represented by the marker point on the patient's head in the camera coordinate system, it is specifically used to:

[0049] Acquiring a head tracking coordinate system image of L markers on the patient's head in the head tracking coordinate system, where the L markers can reflect a spatial shape of the head tracking coordinate system, where L is a positive integer greater than or equal to 3, and at least three of the L markers are not coplanar or collinear;

[0050] A second transformation relationship between the head tracking coordinate system and the camera coordinate system is determined according to the L marking points of the head tracking coordinate system image.

[0051] Optionally, the system further includes:

[0052] an acquisition unit, configured to acquire marker point images of the L marker points on the patient's head;

[0053] An analyzing unit is configured to analyze the running trajectory of the head tracking coordinate system reflected in the marker point image to obtain motion-compensated coordinates of the head tracking coordinate system.

[0054] Optionally, the system further includes:

[0055] The acquisition unit is further configured to acquire a magnetic resonance imaging (MRI) image of the patient's head using a standard template;

[0056] a conversion unit, configured to convert the nuclear magnetic resonance image into a bitmap image;

[0057] A construction unit is configured to construct the three-dimensional virtual head model of the patient's head using the bitmap image.

[0058] Optionally, the system further includes:

[0059] The registration unit is further used to register the three-dimensional virtual head model with the template data of the standard electrode placement method to obtain the 10-20 coordinates on the three-dimensional virtual head model.

[0060] In a third aspect, the present application provides a computer device, comprising:

[0061] Processor, memory, bus, input and output interfaces, wireless network interface;

[0062] The processor is connected to the memory, the input and output interface, and the wireless network interface via a bus;

[0063] The memory stores a program;

[0064] When the processor executes the program stored in the memory, the three-dimensional registration method of the transcranial magnetic stimulation navigation process described in the first aspect is implemented.

[0065] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer storage medium stores instructions, and when the instructions are executed on a computer, the computer executes the three-dimensional registration method of the transcranial magnetic stimulation navigation process as described in the first aspect above.

[0066] In a fifth aspect, the present application provides a computer program product, which, when executed on a computer, enables the computer to perform the three-dimensional registration method of the transcranial magnetic stimulation navigation process as described in the first aspect above.

[0067] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0068] The three-dimensional registration method of the transcranial magnetic stimulation navigation process of the present application determines the first transformation relationship of the probe coordinate system in the camera coordinate system, and determines the second transformation relationship of the head tracking coordinate system represented by the marker point on the patient's head in the camera coordinate system; when the probe tip of the probe coordinate system touches the actual feature point on the patient's head, the coordinates of the actual feature point in the camera coordinate system are calculated using the first transformation relationship, and then the coordinates of the actual feature point in the head tracking coordinate system are calculated using the second transformation relationship, and then the virtual feature points corresponding to the actual feature points in the preset three-dimensional virtual head model are picked up, wherein the three-dimensional virtual head model is located in the three-dimensional virtual head model coordinate system, and the virtual feature points corresponding to the actual feature points are registered to obtain the third transformation relationship between the head tracking coordinate system and the three-dimensional virtual head model coordinate system, and the three-dimensional virtual head model is mapped to the head tracking coordinate system using the third transformation relationship, so that the patient's head and the three-dimensional virtual head model can be registered. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 This is a flow chart of an embodiment of a three-dimensional registration method for the transcranial magnetic stimulation navigation process of the present application;

[0070] Figure 2 This is a flow chart of another embodiment of a three-dimensional registration method for the transcranial magnetic stimulation navigation process of the present application;

[0071] Figure 3 This is a flow chart of another embodiment of a three-dimensional registration method for the transcranial magnetic stimulation navigation process of the present application;

[0072] Figure 4This is a schematic structural diagram of an embodiment of a three-dimensional registration system for the transcranial magnetic stimulation navigation process of the present application;

[0073] Figure 5 This is a schematic diagram of the structure of an embodiment of a computer device of the present application;

[0074] Figure 6 This is a rendering of an embodiment of marking the 10-20 coordinates of a 3D virtual head model of a patient's head with a sphere of 4 mm radius in 3ds Max;

[0075] Figure 7 for Figure 6 A top view of the embodiment effect diagram;

[0076] Figure 8 This is a schematic diagram of the structure of an embodiment of the probe of the present application;

[0077] Figure 9 Schematic diagram of probe images of 6 groups of probe markers in different postures collected by the camera in this application;

[0078] Figure 10 This is a schematic diagram showing the effect of an embodiment in which a positioning reference frame is fixed on a patient's head for simulation purposes. DETAILED DESCRIPTION

[0079] In order to make the purpose, technical solutions and advantages of this application more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0080] As is known to all, there is a close relationship between human mental illness and abnormal cerebral cortical function. For example, studies have shown that depression is related to abnormal function of the prefrontal lobe of the brain. Therefore, accurately providing information on the functional zoning of the cerebral cortex is crucial for transcranial magnetic stimulation treatment. Transcranial magnetic stimulation navigation refers to the process in which the transcranial magnetic stimulation navigation system converts the coordinates of the virtual feature points selected on the specified head model into the navigation coordinates of the magnetic stimulation actuator, so as to realize the magnetic stimulation actuator to magnetically stimulate specific parts of the patient according to the navigation coordinates. The three-dimensional alignment of the transcranial magnetic stimulation navigation process refers to the process of mapping the patient's head with the corresponding preset three-dimensional virtual head model before transcranial magnetic stimulation navigation, that is, the technology of calibrating and matching the three-dimensional virtual head model reconstructed based on the image data in the transcranial magnetic stimulation system with the patient's head. The embodiment of the present application mainly realizes the three-dimensional alignment of the transcranial magnetic stimulation navigation process in the camera coordinate system of the camera.

[0081] See also Figure 1 , an embodiment of the three-dimensional registration method for the transcranial magnetic stimulation navigation process of the present application includes:

[0082] 101. Determine a first transformation relationship between the probe coordinate system and the camera coordinate system.

[0083] This step requires knowing the first transformation relationship of the probe coordinate system where the probe is located in the camera coordinate system of the camera. The camera is preferably an infrared binocular vision camera. For example, the binocular vision system of the infrared binocular vision camera is the MV-VS220 binocular stereo vision system developed by Weishi Intelligent Manufacturing Co., Ltd. This system is a relatively complete system that integrates image acquisition, processing, matching, and measurement, which is convenient for processing image data.

[0084] Specifically, an infrared binocular vision camera is used to obtain a probe image of K marking points of the probe in a probe coordinate system, where the K marking points can reflect the contour shape of the probe, K is a positive integer greater than or equal to 3, and at least 3 of the K marking points are not coplanar and non-collinear. The distance between each pair of marking points on the probe is calculated in the spatial coordinate system constructed by the infrared binocular vision camera, so that a probe coordinate system of the probe can be created. The probe coordinate system is located in the camera coordinate system, and the first transformation relationship of the probe coordinate system in the camera coordinate system can be determined based on the K marking points of the probe image.

[0085] In one possible embodiment, the binocular vision system can use a trained neural network model to identify and classify the shape features composed of K marking points of the probe image, and obtain the first transformation relationship between the probe coordinate system and the camera coordinate system. The training process of the neural network model is a relatively mature existing technology and will not be introduced in detail here.

[0086] For further information, see Figure 8 , Figure 8 A schematic diagram of the structure of a probe is shown. The outline of the probe is in the shape of a "Y", and the four marking points on the probe form a "Y"-shaped outline.

[0087] The first transformation relationship can also be used to calculate the spatial coordinates of the probe tip in the camera coordinate system. For example, please refer to Figure 8 and Figure 9 , a passive optical marker can be fixed at the probe tip position so that the probe tip is exactly at the center of the passive optical marker point. Then, the infrared binocular vision camera is used to obtain the probe image of the four marker points of the probe in the probe coordinate system. The coordinates of the center of the passive optical marker point in the camera coordinate system are the spatial coordinates of the probe tip in the camera coordinate system. Specifically, 6 groups of probe images of the probe in different postures can be collected, and the spatial coordinates of the probe tip position in the camera coordinate system can be calculated respectively. Then, the average value of each group of postures is obtained as the spatial coordinate of the probe tip in the camera coordinate system, and the precise coordinates of the probe tip in different postures in the camera coordinate system are obtained.

[0088] 102. Determine a second transformation relationship between the head tracking coordinate system represented by the marker points on the patient's head and the camera coordinate system.

[0089] This step requires knowing the second transformation relationship between the head tracking coordinate system represented by the marker points on the patient's head and the camera coordinate system of the camera. The camera is preferably an infrared binocular vision camera. For example, the binocular vision system of the infrared binocular vision camera is the MV-VS220 binocular stereo vision system developed by Weishi Intelligent Manufacturing Co., Ltd. It is understandable that since different patients have different head shapes, it is difficult to accurately determine the spatial position of the patient's head in the camera coordinate system directly through the binocular vision system of the camera. In order to solve this technical problem, the method adopted in this step is to use an infrared camera to identify the head tracking coordinate system represented by the marker points on the patient's head. Since these marker points can be fixed to the patient's head in the form of straps through a standard positioning reference frame, the marker points on the positioning reference frame can be well recognized by the binocular vision system of the camera, so that the camera can determine the second transformation relationship between the head tracking coordinate system represented by the marker points on the patient's head and the camera coordinate system. At this time, the camera still does not know the actual spatial position of the patient's head in the head tracking coordinate system, but it has achieved that the camera can determine the position of the head tracking coordinate system represented by the marker points on the patient's head in the camera coordinate system. Please refer to Figure 10 , Figure 10 A schematic diagram showing the effect of simulating an embodiment of a positioning reference frame being fixed on a patient head model. Figure 10 The positioning reference frame (with three infrared reflective balls placed on it) is fixed on the human head model and its relative position with the human head model remains unchanged.

[0090] 103. When the probe tip of the probe coordinate system touches X actual feature points on the patient's head respectively, coordinates of the X actual feature points in the camera coordinate system are calculated using the first transformation relationship.

[0091] After the camera learns the first transformation relationship of the probe coordinate system in the camera coordinate system in step 101, and after the camera learns the second transformation relationship of the head tracking coordinate system represented by the marker points on the patient's head in the camera coordinate system in step 102, this step allows the probe tip of the probe coordinate system to touch the actual feature points on the patient's head, thereby aligning the probe tip with the actual feature points on the patient's head, so as to depict the spatial position of the patient's head in the camera space coordinate system for the camera coordinate system. For example, when the probe tip of the probe coordinate system touches X actual feature points on the patient's head, the coordinates of the X actual feature points in the camera coordinate system are calculated using the first transformation relationship, where X is a positive integer greater than or equal to 3, and at least three of the X actual feature points are non-coplanar and non-collinear. Specifically, the probe tip can be touched to the left ear front point, right ear front point and nose tip of the patient's head in turn to obtain the coordinates of the left ear front point, right ear front point and nose tip in the camera coordinate system. That is, when the probe tip touches the left ear front point, right ear front point and nose tip respectively, the coordinates of the probe tip in the camera coordinate system at that time are the coordinates of the left ear front point, right ear front point and nose tip.

[0092] 104. Calculate the coordinates of the X actual feature points in the head tracking coordinate system using the second transformation relationship.

[0093] After determining the coordinates of the x actual feature points on the patient's head in the camera coordinate system in step 103, since the second transformation relationship between the head tracking coordinate system and the camera coordinate system has already been determined in step 102, and the patient's head is located in the head tracking coordinate system, i.e., the patient's head is in an absolute position in the head tracking coordinate system, the coordinates of the x actual feature points on the patient's head in the head tracking coordinate system can be calculated using the second transformation relationship. For example, the coordinates of the left and right ear points, as well as the tip of the nose, in the head tracking coordinate system can be calculated using the second transformation relationship.

[0094] 105 . Pick X virtual feature points corresponding to the X actual feature points in the preset three-dimensional virtual head model, where the three-dimensional virtual head model is located in a three-dimensional virtual head model coordinate system.

[0095] It is worth noting that the transcranial magnetic stimulation navigation process relies on the spatial coordinates of the preset three-dimensional virtual head model and the spatial coordinates of the patient's head in reality (i.e., the patient's head coordinates in the head tracking coordinate system in the camera coordinate system) to be aligned, so that the three-dimensional virtual head model corresponds to the patient's head coordinates one by one. When the three-dimensional virtual head model specifies transcranial magnetic stimulation for a specific coordinate point, it will command the magnetic stimulation actuator to perform magnetic stimulation on the corresponding specific coordinate point on the patient's head in reality. Based on this understanding, this step requires picking up X virtual feature points in the preset three-dimensional virtual head model for the X actual feature points whose coordinate positions are determined in the camera coordinate system in step 104, where the three-dimensional virtual head model is located in the three-dimensional virtual head model coordinate system. For example, this step can pick up the three coordinate points of the left ear front point, the right ear front point, and the root of the nose in the three-dimensional virtual head model based on the vtkPicker class. VTK (visualization toolkit) is a free, open source software system mainly used for three-dimensional computer graphics, image processing and visualization.

[0096] 106. Register the X virtual feature points corresponding to the X actual feature points respectively to obtain a third transformation relationship between the head tracking coordinate system and the three-dimensional virtual head model coordinate system.

[0097] The X virtual feature points picked in step 105 are respectively registered with the X actual feature points whose coordinate positions are determined in the camera coordinate system in step 104. For example, the registration can be implemented using the Landmark function in VTK to obtain a third transformation relationship between the head tracking coordinate system and the 3D virtual head model coordinate system. Among them, the vtkLandmarkTransform class is one of the more classic registration algorithms in VTK. It is based on marker points and uses linear transformation to minimize the average distance between the two point sets (the patient's head and the three-dimensional virtual head model) after registration. The vtkLandmark algorithm is also relatively simple. The source point set (such as the coordinate set of the three-dimensional virtual head model) and the target point set (such as the coordinate set of the patient's head in the camera coordinate system) are set respectively through the SetSourceLandmarks and SetTargetLandmarks functions. The transformation type is set through SetMode. For example, the SetModeToRigidBody() function sets the registration transformation type to rigid body transformation, and the SetModeToSimilarity() function sets the similarity transformation. The transformation matrix is ​​obtained by setting the GetMatrix() function, which can reduce errors caused by manual operations, shorten treatment time, and reduce surgical risks.

[0098] It can be understood that since X is a positive integer greater than or equal to 3, and at least 3 of the X actual feature points are not coplanar and non-collinear, then after respectively aligning the X virtual feature points corresponding to the X actual feature points, it is equivalent to basically determining the third transformation relationship between the entire three-dimensional virtual head model and the patient's head in the camera coordinate system.

[0099] 107. Map the three-dimensional virtual head model to the head tracking coordinate system using the third transformation relationship, so that the patient's head is aligned with the three-dimensional virtual head model.

[0100] The entire 3D virtual head model can be mapped to the head tracking coordinate system using the third transformation relationship in step 106 , so that the patient's head and the 3D virtual head model are aligned in the camera coordinate system.

[0101] The embodiment of the present application can also realize motion tracking of the head tracking coordinate system through the binocular vision system of the camera, so as to know the real-time movement of the patient's head, provide accurate patient head movement data support for the transcranial magnetic stimulation treatment process, and command the magnetic stimulation actuator to compensate for the movement of the patient's head during operation, thus realizing dynamic and precise magnetic stimulation treatment. Figure 2 , in the above Figure 1 After step 102, another embodiment of the three-dimensional registration method for the transcranial magnetic stimulation navigation process of the present application includes:

[0102] 201. Collect marker images of L markers on the patient's head.

[0103] For example, an infrared binocular vision camera is used to collect in real time the marker point images of L marker points on a positioning reference frame fixed to the patient's head.

[0104] 202. Analyze the running trajectory of the head tracking coordinate system reflected in the marker point image to obtain motion compensation coordinates of the head tracking coordinate system.

[0105] The infrared camera identifies the head tracking coordinate system represented by the markers on the patient's head. Since these markers can be fixed to the patient's head in the form of a strap through a standard positioning reference frame, the markers on the positioning reference frame can be well identified by the camera's binocular vision system. By analyzing the trajectory of the head tracking coordinate system reflected in the marker images between the previous and next moments, the motion compensation coordinates of the head tracking coordinate system are obtained. At this time, the camera does not need to know the actual spatial position of the patient's head in the head tracking coordinate system. It only needs to know the motion compensation coordinates of the head tracking coordinate system, which are the motion compensation coordinates of the patient's head, because the patient's head is in an absolute position relationship in the head tracking coordinate system. This step can use the GPU module in the opencv software to accelerate the image processing process, and then use the CV_TM_CCOEFF_NORMED normalized correlation coefficient matching method to track L markers, achieve the running tracking effect of the patient's head, compensate for the patient's head movement, and make the magnetic stimulation coil of the transcranial magnetic stimulation process in the most suitable stimulation position under the drive of the magnetic stimulation actuator, so as to achieve a better magnetic stimulation treatment effect. The magnetic stimulation actuator of the embodiment of the present application is generally a multi-axis robotic arm, and a magnetic stimulation coil is fixedly installed at the end of the multi-axis robotic arm.

[0106] See also Figure 3 ,exist Figure 1 Before step 105 of the embodiment, the present application needs to construct a three-dimensional virtual head model of the patient's head in advance. Another embodiment of the three-dimensional registration method of the transcranial magnetic stimulation navigation process of the present application includes:

[0107] 301. Use a standard template to acquire an MRI image of the patient's head.

[0108] For example, the standard template uses ICBM152 data, which has an image resolution of 256×256 and a voxel size of 1mm×1mm×1mm. In this step, the standard template is used to acquire an MRI image of the patient's head to obtain standardized MRI image data.

[0109] 302. Convert MRI images into bitmap images.

[0110] For example, in this step, the magnetic resonance imaging data obtained in step 301 is segmented by medical imaging analysis software MRIcron and saved as a bmp picture sequence. The so-called bmp picture sequence refers to a bitmap image.

[0111] 303. Construct a three-dimensional virtual head model of the patient's head using the bitmap image.

[0112] For example, the Bitmap image in step 302 is imported into Simpleware software to construct a three-dimensional virtual head model of the patient's head, and the three-dimensional virtual head model is exported and saved in an STL format.

[0113] 304. Register the three-dimensional virtual head model with the template data of the standard electrode placement method to obtain the 10-20 coordinates on the three-dimensional virtual head model.

[0114] Furthermore, in order to obtain a more accurate three-dimensional virtual head model of the patient's head, a template three-dimensional virtual head model of the standard template can be created, and then the three-dimensional geometry processing system MeshLab is used to align the data of the three-dimensional virtual head model of the patient's head with the template three-dimensional virtual head model to obtain a more accurate three-dimensional virtual head model of the patient's head.

[0115] Then, in this step, Brainstorm software is used to obtain the 10-20 coordinates of the template 3D virtual head model. The so-called 10-20 coordinates refer to the placement coordinates of the standard electrode placement method specified by the International Electroencephalography Society in the 10-20 system electrode placement method. This application uses the 10-20 coordinates to better select the stimulation target for the subsequent transcranial magnetic stimulation coil. Since the data of the patient's 3D virtual head model and the template 3D virtual head model have been aligned, the accurate 10-20 coordinates of the patient's 3D virtual head model can also be obtained. Please refer to Figure 6 as well as Figure 7 In one embodiment of the present application, a 4mm radius sphere is used in 3ds Max to mark the 10-20 coordinates of a 3D virtual head model of a patient's head. The 10-20 system facilitates the operator's positioning of the target area on the patient's head, allowing for better control of the magnetic stimulation actuator's magnetic stimulation coils to treat the target area.

[0116] The above embodiment describes the three-dimensional registration method of the transcranial magnetic stimulation navigation process of the present application. The following describes the three-dimensional registration system of the transcranial magnetic stimulation navigation process of the present application. Figure 4 , an embodiment of a three-dimensional registration system for a transcranial magnetic stimulation navigation procedure includes:

[0117] A first determining unit 401 is used to determine a first transformation relationship between the probe coordinate system and the camera coordinate system;

[0118] A second determining unit 402 is configured to determine a second transformation relationship between the head tracking coordinate system represented by the marker points on the patient's head and the camera coordinate system;

[0119] a first calculation unit 403 configured to calculate, when the probe tip in the probe coordinate system touches X actual feature points on the patient's head respectively, coordinates of the X actual feature points in the camera coordinate system using the first transformation relationship, where X is a positive integer greater than or equal to 3, and at least three of the X actual feature points are not coplanar or collinear;

[0120] A second calculation unit 404 is configured to calculate the coordinates of the X actual feature points in the head tracking coordinate system using the second transformation relationship;

[0121] A picking unit 405 is configured to pick X virtual feature points corresponding to the X actual feature points in a preset three-dimensional virtual head model, wherein the three-dimensional virtual head model is located in a three-dimensional virtual head model coordinate system;

[0122] a registration unit 406 for respectively registering the X virtual feature points corresponding to the X actual feature points to obtain a third transformation relationship between the head tracking coordinate system and the three-dimensional virtual head model coordinate system;

[0123] The mapping unit 407 is configured to map the three-dimensional virtual head model to the head tracking coordinate system using the third transformation relationship, so that the patient's head is registered with the three-dimensional virtual head model.

[0124] Optionally, when the first determining unit 401 determines the first transformation relationship between the probe coordinate system and the camera coordinate system, it is specifically used to:

[0125] Acquire a probe image of K markers of the probe in the probe coordinate system, wherein the K markers can reflect the contour shape of the probe, wherein K is a positive integer greater than or equal to 3, and at least three of the K markers are not coplanar and not collinear;

[0126] The first transformation relationship between the probe coordinate system and the camera coordinate system is determined according to the K marking points of the probe image.

[0127] Optionally, K is equal to 4, the contour of the probe is in a "Y" shape, and the four marking points form the "Y"-shaped contour. When the first determination unit 401 determines the first transformation relationship of the probe coordinate system in the camera coordinate system according to the K marking points of the probe image, it is specifically used to:

[0128] The shape features composed of the K marking points of the probe image are identified and classified by a trained neural network model to obtain the first transformation relationship between the probe coordinate system and the camera coordinate system.

[0129] Optionally, the K marking points include a probe tip of the probe, and the system further includes:

[0130] The third calculation unit 408 is configured to calculate the spatial coordinates of the probe tip in the camera coordinate system by using the first transformation relationship.

[0131] Optionally, when the second determining unit 402 determines the second transformation relationship of the head tracking coordinate system represented by the marker point on the patient's head in the camera coordinate system, it is specifically configured to:

[0132] Acquiring a head tracking coordinate system image of L markers on the patient's head in the head tracking coordinate system, where the L markers can reflect a spatial shape of the head tracking coordinate system, where L is a positive integer greater than or equal to 3, and at least three of the L markers are not coplanar or collinear;

[0133] A second transformation relationship between the head tracking coordinate system and the camera coordinate system is determined according to the L marking points of the head tracking coordinate system image.

[0134] Optionally, the system further includes:

[0135] An acquisition unit 409 is configured to acquire marker point images of the L marker points on the patient's head;

[0136] The analyzing unit 410 is configured to analyze the running trajectory of the head tracking coordinate system reflected in the marker image to obtain motion-compensated coordinates of the head tracking coordinate system.

[0137] Optionally, the system further includes:

[0138] The acquisition unit 409 is further configured to acquire the MRI image of the patient's head using a standard template;

[0139] a conversion unit 411, configured to convert the MRI image into a bitmap image;

[0140] The constructing unit 412 is configured to construct the three-dimensional virtual head model of the patient's head using the bitmap image.

[0141] Optionally, the system further includes:

[0142] The registration unit 413 is further configured to register the three-dimensional virtual head model with the template data of the standard electrode placement method to obtain 10-20 coordinates on the three-dimensional virtual head model.

[0143] The three-dimensional registration system of the transcranial magnetic stimulation navigation process of the embodiment of the present application performs the same operations as the aforementioned Figure 1 、 Figure 2 as well as Figure 3 The operations performed in the embodiment are similar and will not be described in detail here.

[0144] The following describes the computer device of the embodiment of the present application. Figure 5 , an embodiment of the computer device in the embodiment of the present application includes:

[0145] The computer device 500 may include one or more processors (central processing units, CPU) 501 and a memory 502, in which one or more applications or data are stored. The memory 502 is a volatile storage or a persistent storage. The program stored in the memory 502 may include one or more modules, each of which may include a series of instruction operations on the computer device. Furthermore, the processor 501 may be configured to communicate with the memory 502 and execute a series of instruction operations in the memory 502 on the computer device 500. The computer device 500 may also include one or more wireless network interfaces 503, one or more input and output interfaces 504, and / or one or more operating systems, such as Windows Server, Mac OS, Unix, Linux, FreeBSD, etc. The processor 501 may execute the aforementioned Figures 1 to 3 The operations performed in the illustrated embodiment will not be described in detail here.

[0146] In the several embodiments provided in the embodiments of the present application, those skilled in the art should understand that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the unit is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0147] In addition, the functional units in the various embodiments of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present application. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0148] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements or improvements made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A three-dimensional registration method for transcranial magnetic stimulation navigation, characterized in that: include: Determine the first transformation relationship between the probe coordinate system and the camera coordinate system; Determine a second transformation relationship between the head tracking coordinate system represented by the marker points on the patient's head and the camera coordinate system; When the probe tip of the probe coordinate system touches X actual feature points on the patient's head respectively, coordinates of the X actual feature points in the camera coordinate system are calculated using the first transformation relationship, where X is a positive integer greater than or equal to 3, and at least three of the X actual feature points are not coplanar or collinear; Calculating the coordinates of the X actual feature points in the head tracking coordinate system using the second transformation relationship; Picking X virtual feature points corresponding to the X actual feature points in a preset three-dimensional virtual head model, wherein the three-dimensional virtual head model is located in a three-dimensional virtual head model coordinate system; respectively registering X virtual feature points corresponding to the X actual feature points to obtain a third transformation relationship between the head tracking coordinate system and the three-dimensional virtual head model coordinate system; Mapping the three-dimensional virtual head model to the head tracking coordinate system using the third transformation relationship, so that the patient's head is aligned with the three-dimensional virtual head model; Based on the X virtual feature points corresponding to the X actual feature points, a linear transformation is performed to minimize the average distance between the registered patient head and the three-dimensional virtual head model; Determining the first transformation relationship of the probe coordinate system in the camera coordinate system includes: Acquire a probe image of K markers of the probe in the probe coordinate system, wherein the K markers can reflect the contour shape of the probe, wherein K is a positive integer greater than or equal to 3, and at least three of the K markers are not coplanar and not collinear; Determining the first transformation relationship of the probe coordinate system in the camera coordinate system according to the K marking points of the probe image; The K is equal to 4, the contour of the probe is in a Y shape, and the four marking points form the Y-shaped contour. The first transformation relationship of the probe coordinate system in the camera coordinate system is determined based on the K marking points of the probe image, including: Recognizing and classifying shape features composed of the K marking points of the probe image through a trained neural network model to obtain the first transformation relationship between the probe coordinate system and the camera coordinate system; The K marking points include a probe tip of the probe. After obtaining the first transformation relationship between the probe coordinate system and the camera coordinate system, the method further includes: Calculating the spatial coordinates of the probe tip in the camera coordinate system using the first transformation relationship; The second transformation relationship of determining the head tracking coordinate system represented by the marker points on the patient's head in the camera coordinate system includes: Acquiring a head tracking coordinate system image of L markers on the patient's head in the head tracking coordinate system, where the L markers can reflect a spatial shape of the head tracking coordinate system, where L is a positive integer greater than or equal to 3, and at least three of the L markers are not coplanar or collinear; A second transformation relationship between the head tracking coordinate system and the camera coordinate system is determined according to the L marking points of the head tracking coordinate system image.

2. The three-dimensional registration method according to claim 1, characterized in that: After determining the second transformation relationship between the head tracking coordinate system and the camera coordinate system, the method further includes: Acquire marker point images of the L marker points on the patient's head; The running trajectory of the head tracking coordinate system reflected in the marker point image is analyzed to obtain motion compensation coordinates of the head tracking coordinate system.

3. The three-dimensional registration method according to claim 1, characterized in that: Before picking X virtual feature points corresponding to the X actual feature points in the preset three-dimensional virtual head model, the method further includes: Acquiring an MRI image of the patient's head using a standard template; converting the magnetic resonance image into a bitmap image; The three-dimensional virtual head model of the patient's head is constructed using the bitmap image.

4. The three-dimensional registration method according to claim 3, characterized in that: After constructing the three-dimensional virtual head model of the patient using the bitmap image, the method further includes: The three-dimensional virtual head model is registered with the template data of the standard electrode placement method to obtain the 10-20 coordinates on the three-dimensional virtual head model.

5. A three-dimensional registration system for transcranial magnetic stimulation navigation, characterized in that: include: A first determining unit, configured to determine a first transformation relationship between the probe coordinate system and the camera coordinate system; a second determining unit, configured to determine a second transformation relationship between the head tracking coordinate system represented by the marker points on the patient's head and the camera coordinate system; a first calculation unit, configured to calculate, when the probe tip of the probe coordinate system touches X actual feature points on the patient's head respectively, coordinates of the X actual feature points in the camera coordinate system using the first transformation relationship, where X is a positive integer greater than or equal to 3, and at least three of the X actual feature points are not coplanar or collinear; A second calculation unit is configured to calculate the coordinates of the X actual feature points in the head tracking coordinate system using the second transformation relationship; a picking unit, configured to pick up X virtual feature points corresponding to the X actual feature points in a preset three-dimensional virtual head model, wherein the three-dimensional virtual head model is located in a three-dimensional virtual head model coordinate system; a registration unit, configured to respectively register the X virtual feature points corresponding to the X actual feature points, to obtain a third transformation relationship between the head tracking coordinate system and the three-dimensional virtual head model coordinate system; a mapping unit, configured to map the three-dimensional virtual head model to the head tracking coordinate system using the third transformation relationship, so that the patient's head is registered with the three-dimensional virtual head model; Based on the X virtual feature points corresponding to the X actual feature points, a linear transformation is performed to minimize the average distance between the registered patient head and the three-dimensional virtual head model; When the first determining unit determines the first transformation relationship between the probe coordinate system and the camera coordinate system, it is specifically used to: Acquire a probe image of K markers of the probe in the probe coordinate system, wherein the K markers can reflect the contour shape of the probe, wherein K is a positive integer greater than or equal to 3, and at least three of the K markers are not coplanar and not collinear; Determining the first transformation relationship of the probe coordinate system in the camera coordinate system according to the K marking points of the probe image; The K is equal to 4, the contour of the probe is in a Y shape, and the four marking points form the Y-shaped contour. When the first determining unit determines the first transformation relationship of the probe coordinate system in the camera coordinate system based on the K marking points of the probe image, it is specifically used to: Recognizing and classifying shape features composed of the K marking points of the probe image through a trained neural network model to obtain the first transformation relationship between the probe coordinate system and the camera coordinate system; The K marking points include a probe tip of the probe, and the system further includes: a third calculation unit, configured to calculate the spatial coordinates of the probe tip in the camera coordinate system using the first transformation relationship; When the second determining unit determines the second transformation relationship between the head tracking coordinate system represented by the marker point on the patient's head and the camera coordinate system, it is specifically used to: Acquiring a head tracking coordinate system image of L markers on the patient's head in the head tracking coordinate system, where the L markers can reflect a spatial shape of the head tracking coordinate system, where L is a positive integer greater than or equal to 3, and at least three of the L markers are not coplanar or collinear; A second transformation relationship between the head tracking coordinate system and the camera coordinate system is determined according to the L marking points of the head tracking coordinate system image.

6. A computer device, characterized in that: include: Processor, memory, bus, input and output interfaces, wireless network interface; The processor is connected to the memory, the input and output interface, and the wireless network interface via a bus; The memory stores a program; When the processor executes the program stored in the memory, the three-dimensional registration method for the transcranial magnetic stimulation navigation process according to any one of claims 1 to 4 is implemented.

Images