Individualized wearable transcranial magnetic stimulation therapy assisted positioning method

Abstract

The invention discloses an individualized wearable transcranial magnetic stimulation treatment assisted positioning method, which comprises the following steps: based on T1 weighted structure imaging, obtaining a scalp surface, a skull surface and a cerebral cortex surface consistent with an individual head anatomy through surface reconstruction; on the basis of resting state functional magnetic resonance imaging, whole brain function connection analysis is executed to obtain an individualized cortex target spot; identifying anatomical mark points on the scalp surface and constructing a scalp curved surface coordinate system, and projecting the individualized cortex target spot from the cerebral cortex surface to the scalp surface to obtain an individualized scalp target spot position; selecting a scalp curved surface above an ear plane in the scalp curved surface coordinate system as a hood base area, making a three-dimensional model of the individualized positioning hood, and reserving a coil placement area corresponding to the individualized scalp target according to the position of the individualized scalp target; an individualized positioning hood entity is manufactured according to the three-dimensional model, and individualized accurate positioning, repeated placement and standardized operation of transcranial magnetic stimulation are achieved.

Description

Technical Field

[0001] This invention belongs to the field of neuromodulation and medical engineering technology, specifically relating to a personalized wearable transcranial magnetic stimulation therapy auxiliary positioning method. Background Technology

[0002] Transcranial magnetic stimulation (rTMS) is a non-invasive neuromodulation technique based on the principle of electromagnetic induction. It involves placing stimulation coils on the scalp to generate a rapidly changing magnetic field, thereby inducing an electric current within the brain tissue to excite or inhibit the cortex and related neural circuits. Currently, repetitive transcranial magnetic stimulation (rTMS) is widely used in the treatment of various neuropsychiatric disorders, including depression, autism, and attention deficit hyperactivity disorder (ADHD).

[0003] However, in current clinical practice, the efficacy of transcranial magnetic stimulation (TMS) varies significantly among individuals. One important reason is the lack of standardization in target localization and the high dependence of coil placement on operator experience. Commonly used 5 cm rules and experience-based localization methods based on 10-20 EEG leads cannot fully reflect individual anatomical and functional connectivity differences, leading to deviations between the stimulation site and the actual pathologically relevant brain regions. Furthermore, it is difficult to maintain consistency in coil placement position and angle across multiple treatment sessions, on different treatment days, and with different operators, further affecting the stability of efficacy.

[0004] To improve spatial accuracy, various transcranial magnetic stimulation (TMS) navigation schemes based on brain network maps and image navigation have been proposed in existing technologies. For example, by acquiring individual cranial MRI (magnetic resonance imaging) data and registering it with a partition template (such as Brainnetome Atlas), the MNI coordinates of cranial functional partitions and cortical target points are obtained. Then, combined with a neuronavigation system (such as optical navigation), the coil placement is guided in real time during each treatment. Based on this, a target mapping model or auxiliary positioning device is established.

[0005] While the aforementioned methods can improve the accuracy of target localization, they heavily rely on expensive navigation equipment and specialized operation. Their application in multi-center settings and routine clinical use still suffers from high costs, complex procedures, and repetitiveness dependent on operator experience. Therefore, there is an urgent need for a technical solution that, based on fully utilizing individual MRI / functional imaging to determine cortical targets, can directly construct a stable coordinate system on the individual's scalp geometry and automatically generate a personalized positioning headgear that conforms to the head shape. This solution would achieve precise and repetitive localization and standardized operation of transcranial magnetic stimulation targets without relying on complex navigation systems. Summary of the Invention

[0006] To address the aforementioned shortcomings in existing technologies, this invention provides an individualized wearable transcranial magnetic stimulation (TMS) therapy auxiliary positioning method that solves the problem that coil placement in existing TMS therapy is highly dependent on the doctor's experience and makes it difficult to ensure consistent spatial positioning between different treatment courses and different operators.

[0007] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows: a personalized wearable transcranial magnetic stimulation therapy-assisted positioning method, comprising the following steps: S1. Acquire resting-state functional magnetic resonance imaging (fMRI) and T1-weighted structural imaging of the patient, and preprocess the resting-state fMRI. S2. Based on T1-weighted structural imaging, surface reconstruction is used to obtain the scalp surface, skull surface and cerebral cortex surface that are consistent with the individual head anatomy. S3. Based on preprocessed resting-state functional magnetic resonance imaging, using disease-related deep nuclei as seed points, perform whole-brain functional connectivity analysis, and screen out the regions with the strongest functional connectivity with the seed points in the cortical region to obtain individualized cortical targets. S4. Identify anatomical landmarks on the scalp surface, construct a scalp surface coordinate system based on the geodesic distance between the anatomical landmarks, and project individualized cortical target points from the cerebral cortex surface to the scalp surface to obtain the individualized scalp target point locations. S5. In the scalp surface coordinate system, select the scalp surface above the ear plane as the base area of ​​the head cover, expand along the outer normal vector of this area to form a three-dimensional model of the individualized positioning head cover, and reserve the coil placement area corresponding to the individualized scalp target point according to the individualized scalp target point position. S6. Based on the three-dimensional model, create an individualized positioning headgear for the repeated and standardized placement of the transcranial magnetic stimulation coil in terms of spatial position and orientation.

[0008] Furthermore, in S1, the preprocessing methods for resting-state functional magnetic resonance imaging include: removing the first 10 time points, time-level correction, spatial correction, standardization, smoothing to delinear drift, bandpass filtering, regression covariates, and cubic spline interpolation.

[0009] Furthermore: S2 specifically refers to: S21. Model and segment the scalp surface, skull surface, and cerebral cortex surface based on T1-weighted structural imaging. The specific expressions for modeling and segmenting the scalp surface, skull surface, and cerebral cortex surface are as follows: In the formula, voxels Organization category label, Used for voxels Marked as At that time, the measurement of "voxels" "Actual grayscale value" and "Organization category" The degree of inconsistency between the "gray intensity models" It is a neighborhood set between voxels. For smoothing terms, voxels within the neighborhood Organization category label, To be related to voxels Forming neighborhood pairs Another voxel index, For in the marked field The total energy function under the following conditions These are the weighting coefficients for the smoothing term, used to balance the relative contributions of the data term and the spatial smoothing term; S22. Reconstruct the scalp surface, skull surface, and cerebral cortex surface based on the modeling and segmentation results. The specific expressions for reconstructing the scalp surface, skull surface, and cerebral cortex surface are as follows: In the formula, The total energy function reflects the degree of deviation between the current surface shape and the desired solution. Let be the set of three-dimensional coordinates of all vertices on the surface to be reconstructed. Energy is the data term used to measure the consistency between the reconstructed surface and the grayscale / boundary information of the original magnetic resonance image. The smoothing energy is used to measure the local smoothness of a surface. These are the weighting coefficients for the data items, used to adjust the degree of importance attached to the image data. This is the weighting coefficient for the smoothing term, used to adjust the importance of surface smoothing constraints.

[0010] Furthermore: In S3, the expression for performing whole-brain functional connectivity analysis is specifically as follows: In the formula, For the first i The strength of the functional connectivity between voxels and seed sites t For time point indexing, T This represents the total number of time points in the resting-state magnetic resonance time series. For at a certain point in time t At that time, the time series mean of the BOLD signal in deep nuclei, The seed time series is the average value over the entire time period. , For at a certain point in time t At that time, the first i Time series values ​​of the BOLD signal of individual voxels For the firsti The mean of the individual time series. .

[0011] Furthermore, S4 includes the following sub-steps: S41. Determine the grid vertex coordinates of anatomical landmarks on the 3D grid of the scalp. The anatomical landmarks include the root of the nose, the external occipital protuberance, the left preauricular point, and the right preauricular point. S42. Calculate the geodesic distance from any scalp vertex to the anatomical landmark using the surface geodesic distance algorithm, and obtain the geodesic distances of the root of the nose, external occipital protuberance, left preauricular point and right preauricular point; S43. Using normalization, construct the horizontal coordinate based on the difference in geodesic distance between the left and right preauricular points, and construct the vertical coordinate based on the difference in geodesic distance between the root of the nose and the external occipital protuberance, forming a scalp surface coordinate system constrained by four anatomical points. S44. In the scalp surface coordinate system, the individualized cortical target point is projected onto the scalp surface along the radial or shortest path to obtain a unique scalp target point coordinate, which serves as the individualized scalp target point position to guide coil placement.

[0012] Furthermore: S5 specifically refers to: Based on the area above the ear plane determined by the line connecting the left and right preauricular points, the upper half of the scalp surface of the top of the head and the top of the forehead is selected as the base area of ​​the headgear in the scalp surface coordinate system. A thin shell structure is formed by extrusion along the outer normal vector to generate a three-dimensional model of the individualized positioning headgear that matches the patient's head shape. According to the individualized scalp target point position, a coil placement area corresponding to the individualized scalp target point is reserved on the surface of the individualized positioning headgear.

[0013] Furthermore: S6 specifically refers to: The 3D model is exported as a 3D printable data format, and an individualized positioning headgear is made by 3D printing. The outer surface of the headgear retains positioning marks and strap fixation structures corresponding to the target points, thereby enabling the repeated placement and standardized operation of the transcranial magnetic stimulation coil spatial position and orientation among different operators.

[0014] The beneficial effects of this invention are as follows: (1) This invention provides an individualized wearable transcranial magnetic stimulation (TMS) therapy auxiliary positioning method. In view of the problem that the placement of coils in existing TMS therapy is highly dependent on the doctor's experience and it is difficult to ensure the consistency of spatial position between different treatment courses and different operators, this invention obtains individualized scalp target points in the individual space through whole brain functional connectivity analysis, and constructs a physical headgear based on the three-dimensional model of the individualized positioning headgear, realizing a complete mapping chain from the individual brain to the individual scalp and then to the individual headgear. Without relying on a complex optical navigation system, it realizes individualized precise positioning, repeated placement and standardized operation of TMS, reduces manual positioning error, improves the spatial accuracy of TMS therapy and the consistency between treatment courses, and has good clinical promotion value and universal applicability.

[0015] (2) The present invention establishes a scalp surface coordinate system by using the nasal root point, external occipital protuberance, left preauricular point and right preauricular point, so that the target point position is not only repeatable in space, but also has clear coordinate parameters in the coordinate system, which can be compared and recorded across time, across operators and even across devices.

[0016] (3) The individualized headgear designed in this invention provides stable geometric constraints, which can significantly reduce target point deviation caused by slight changes in head position or differences in operator experience. This helps to improve the spatial consistency of transcranial magnetic stimulation, greatly reduce the dependence on optical navigation system and operating costs in subsequent treatments, and improve overall efficacy and repeatability. Attached Figure Description

[0017] Figure 1 This is a flowchart of an individualized wearable transcranial magnetic stimulation therapy auxiliary positioning method according to the present invention.

[0018] Figure 2 This invention creates a standard template for the dorsolateral prefrontal cortex based on actual assumptions.

[0019] Figure 3 This invention uses a surface reconstruction algorithm to obtain scalp, skull, and cerebral cortex surface maps that are consistent with the anatomy of an individual's head.

[0020] Figure 4 A coordinate system diagram of the scalp surface is constructed for this invention.

[0021] Figure 5 This is a map showing the individualized scalp target location obtained by the present invention.

[0022] Figure 6 This invention generates an individualized positioning headgear diagram that matches the patient's head shape. Detailed Implementation

[0023] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0024] like Figure 1 As shown, in one embodiment of the present invention, a personalized wearable transcranial magnetic stimulation therapy-assisted positioning method includes the following steps: S1. Acquire resting-state functional magnetic resonance imaging (fMRI) and T1-weighted structural imaging of the patient, and preprocess the resting-state fMRI. S2. Based on T1-weighted structural imaging, surface reconstruction is used to obtain the scalp surface, skull surface and cerebral cortex surface that are consistent with the individual head anatomy. S3. Based on preprocessed resting-state functional magnetic resonance imaging, using disease-related deep nuclei as seed points, perform whole-brain functional connectivity analysis, and screen out the regions with the strongest functional connectivity with the seed points in the cortical region to obtain individualized cortical targets. S4. Identify anatomical landmarks on the scalp surface, construct a scalp surface coordinate system based on the geodesic distance between the anatomical landmarks, and project individualized cortical target points from the cerebral cortex surface to the scalp surface to obtain the individualized scalp target point locations. S5. In the scalp surface coordinate system, select the scalp surface above the ear plane as the base area of ​​the head cover, expand along the outer normal vector of this area to form a three-dimensional model of the individualized positioning head cover, and reserve the coil placement area corresponding to the individualized scalp target point according to the individualized scalp target point position. S6. Based on the three-dimensional model, create an individualized positioning headgear for the repeated and standardized placement of the transcranial magnetic stimulation coil in terms of spatial position and orientation.

[0025] In S1, the preprocessing methods for resting-state functional magnetic resonance imaging include: removing the first 10 time points, time-level correction, spatial correction, standardization, smoothing to delinear drift, bandpass filtering (0.01-0.1Hz), regression covariates (white matter, cerebrospinal fluid, 24 head motion parameters), and cubic spline interpolation (scrubbing).

[0026] S2 specifically refers to: S21. Model and segment the scalp surface, skull surface, and cerebral cortex surface based on T1-weighted structural imaging. The specific expressions for modeling and segmenting the scalp surface, skull surface, and cerebral cortex surface are as follows: In the formula, voxels Organization category label, Used for voxels Marked as At that time, the measurement of "voxels" "Actual grayscale value" and "Organization category" The degree of inconsistency between the "gray intensity models" It is a neighborhood set between voxels. For smoothing terms, voxels within the neighborhood Organization category label, To be related to voxels Forming neighborhood pairs Another voxel index, For in the marked field The total energy function under the condition of minimizing To obtain the optimal segmentation result, These are the weighting coefficients for the smoothing term, used to balance the relative contributions of the data term and the spatial smoothing term; In this embodiment, through the total energy function Perform modeling and segmentation to obtain The resulting labeled field clearly defines the voxel sets and their interfaces for the scalp, skull, and cerebral cortex. Subsequent surface reconstruction involves fitting smooth surfaces onto these interfaces. Therefore, the segmentation results provide target boundaries and constraints for the reconstruction of the scalp, skull, and cerebral cortex surfaces.

[0027] S22. Reconstruct the scalp surface, skull surface, and cerebral cortex surface based on the modeling and segmentation results. The specific expressions for reconstructing the scalp surface, skull surface, and cerebral cortex surface are as follows: In the formula, The total energy function reflects the degree of deviation between the current surface shape and the desired solution. Energy is the data term used to measure the consistency between the reconstructed surface and the grayscale / boundary information of the original magnetic resonance image. The smoothing energy is used to measure the local smoothness of a surface. These are the weighting coefficients for the data items, used to adjust the degree of importance attached to the image data. These are the weighting coefficients for the smoothing term, used to adjust the importance of surface smoothing constraints. Let be the set of three-dimensional coordinates of all vertices on the surface to be reconstructed. Indicates the first The position of each vertex , and For the first The X-axis, Y-axis, and Z-axis coordinates of each vertex.

[0028] In this embodiment, the set of vertices representing the geometry of the scalp surface, skull surface, and cerebral cortex surface is denoted as s , with total energy function For the objective function, with respect to the objective variable s Solving the energy minimization problem yields the optimal vertex position. s * This refers to the three-dimensional reconstruction results of the scalp surface, skull surface, and cerebral cortex surface.

[0029] In S3, whole-brain functional connectivity analysis is used to analyze the time-series mean of the BOLD signal extracted from deep target nuclei in individual space by resting-state functional magnetic resonance imaging (fMRI). The correlation between this time-series mean and the BOLD signal time-series values ​​of voxels in various cortical layers of the whole brain is calculated to obtain the functional connectivity strength between voxels and seed points. Regions with significant functional connectivity are screened based on statistical significance to determine individualized cortical target points. The specific expression for performing whole-brain functional connectivity analysis is as follows: In the formula, For the first i The strength of the functional connectivity between voxels and seed sites t For time point indexing, T This represents the total number of time points in the resting-state magnetic resonance time series. For at a certain point in time t At that time, the time series mean of the BOLD signal in deep nuclei (e.g., the amygdala), The seed time series is the average value over the entire time period. , For at a certain point in time t At that time, the first i Time series values ​​of the BOLD signal of individual voxels For the first i The mean of the individual time series. .

[0030] S4 includes the following steps: S41. Determine the grid vertex coordinates of anatomical landmarks on the 3D scalp grid. Anatomical landmarks include the root of the nose (NAS), external occipital protuberance (INI), left preauricular point (LPA), and right preauricular point (RPA). S42. Calculate the geodesic distance from any scalp vertex to the anatomical landmark using the surface geodesic distance algorithm, and obtain the geodesic distances of the root of the nose, external occipital protuberance, left preauricular point and right preauricular point; S43. Using normalization, construct the horizontal coordinate u based on the geodesic distance difference between the left and right preauricular points, and construct the vertical coordinate v based on the geodesic distance difference between the nasal root point and the external occipital protuberance, forming a scalp surface coordinate system (u,v) constrained by four anatomical points. S44. In the scalp surface coordinate system, the individualized cortical target point is projected onto the scalp surface along the radial or shortest path to obtain a unique scalp target point coordinate, which serves as the individualized scalp target point position to guide coil placement.

[0031] S5 specifically refers to: Based on the area above the ear plane defined by the line connecting the left and right preauricular points, the upper half of the scalp surface at the top of the head and forehead is selected as the base area of ​​the headgear in the scalp surface coordinate system. A thin shell structure is formed by extrusion along the outer normal vector to generate a three-dimensional model of an individualized positioning headgear that matches the patient's head shape. According to the individualized scalp target point position, a coil placement area corresponding to the individualized scalp target point is reserved on the surface of the individualized positioning headgear so as to rapidly form the individualized positioning headgear through 3D printing for repeated coil positioning and fixation in clinical transcranial magnetic stimulation therapy.

[0032] S6 specifically refers to: The 3D model is exported as a 3D printable data format, and an individualized positioning headgear is made by 3D printing. The outer surface of the headgear retains positioning marks and strap fixation structures corresponding to the target points, thereby enabling the repeated placement and standardized operation of the transcranial magnetic stimulation coil spatial position and orientation among different operators.

[0033] In this embodiment, the individualized positioning headgear is made by 3D printing material. The inner surface of the headgear fits closely to the curved surface of the individual's scalp, while the outer surface of the headgear retains positioning marks and strap fixation structures corresponding to the target points.

[0034] To verify the effectiveness of the present invention, this embodiment uses the amygdala as the seed point to construct individualized target points for the DLPFC (dorsolateral prefrontal cortex) and generate an individualized positioning headgear.

[0035] Taking an autistic patient as an example, the bilateral amygdala was selected as the deep nucleus seed region. Whole-brain functional connectivity was calculated in the individual space to identify the cortical target point within the DLPFC that has the strongest functional connectivity with the amygdala. Based on this target point, a personalized positioning headgear was generated to fit the patient's head shape. The standard template for the DLPFC is as follows: Figure 2 As shown, the standard template of DLPFC is registered to the individual space, and the region with the strongest functional connection with the amygdala in the DLPFC region is found as the precise target point for each patient. Based on the individual's structural imaging, structural images with target markers are generated.

[0036] Step 1: Obtain resting-state functional magnetic resonance imaging (fMRI) and high-resolution T1-weighted structural imaging of the patient. Resting-state fMRI is used for subsequent functional connectivity analysis, and T1-weighted structural imaging is used to reconstruct individualized head and brain anatomy. Step two involves performing routine preprocessing on resting-state functional magnetic resonance imaging (fMRI) and high-resolution T1-weighted structural imaging (T1-weighted structural imaging), and then reconstructing the patient's scalp surface, skull surface, and cerebral cortex surface based on the T1-weighted structural imaging. The constructed results are shown below. Figure 3 As shown, a surface matching the actual shape of the individual's head is obtained.

[0037] Step 3: In the standard space, the amygdala is predefined as a deep seed region, and the amygdala template is registered to the patient's individual space; the average time series of the amygdala is extracted in resting-state functional magnetic resonance imaging, and the functional connectivity is calculated with each location in the whole brain cortex. In the pre-set candidate region of the dorsolateral prefrontal cortex, the cortical point with the highest functional connectivity with the amygdala is selected as the individualized cortical target point for the patient, and its three-dimensional coordinates on the cortical surface are recorded. Step four involves automatically identifying four anatomical landmarks on the reconstructed scalp surface: the nasal root point, the external occipital protuberance, and the left and right preauricular points. Using these four points as anchors, a two-dimensional scalp surface coordinate system is constructed based on the geodesic distance relationships on the scalp surface, such as... Figure 4 As shown, normalized left-right and front-back direction parameters are assigned to any point on the scalp; Step 5: Based on the individualized cortical target points obtained in Step 3, emit a ray in the radial direction from the geometric center of the head towards the target point, intersecting the ray with the scalp surface to obtain the initial scalp target point, and calculate its u and v parameters in the scalp surface coordinate system. If necessary, fine-tune the initial position in the (u,v) space of the scalp surface coordinate system to obtain the final individualized scalp target point position. The obtained individualized scalp target point position is shown below. Figure 5 As shown, this is used to guide the placement of the coil; Step six: Determine the ear plane by connecting the front points of the left and right ears, and divide the scalp surface into areas above and below the ear plane; select the scalp surface above the ear plane as the base area of ​​the headgear, and use a surface coordinate system to limit the front and back range to cover the key areas of the forehead and top of the head. Expand the outer normal vector of this area uniformly by 2-3 mm to form a thin shell structure that fits the head shape, thus obtaining a three-dimensional model of the individualized positioning headgear; Step 7: Mark the coil placement area on the outer surface of the headgear corresponding to the scalp target points obtained in Step 5, and indicate the coil orientation using the direction of the target point connection line; export the 3D model of the headgear into a 3D printable data format, and print a solid headgear using suitable materials. The solid headgear is shown below. Figure 6 As shown, this demonstrates the rapid, repetitive, and standardized placement of auxiliary coils used in clinical transcranial magnetic stimulation therapy.

[0038] In this embodiment, the amygdala is used as the seed region and the point with the strongest functional connectivity in the dorsolateral prefrontal cortex is selected as the target point for illustrative purposes only. The present invention can also select other deep nuclei or cortical networks as seed regions according to different diseases, obtain corresponding individualized cortical target points and generate positioning headgear under the same process.

[0039] In the description of this invention, the above are merely preferred embodiments and are not intended to limit the scope of protection of this invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A personalized wearable transcranial magnetic stimulation (TMS) therapy-assisted positioning method, characterized in that, Includes the following steps: S1. Acquire resting-state functional magnetic resonance imaging (fMRI) and T1-weighted structural imaging of the patient, and preprocess the resting-state fMRI. S2. Based on T1-weighted structural imaging, surface reconstruction is used to obtain the scalp surface, skull surface and cerebral cortex surface that are consistent with the individual head anatomy. S3. Based on preprocessed resting-state functional magnetic resonance imaging, using disease-related deep nuclei as seed points, perform whole-brain functional connectivity analysis, and screen out the regions with the strongest functional connectivity with the seed points in the cortical region to obtain individualized cortical targets. S4. Identify anatomical landmarks on the scalp surface, construct a scalp surface coordinate system based on the geodesic distance between the anatomical landmarks, and project individualized cortical target points from the cerebral cortex surface to the scalp surface to obtain the individualized scalp target point locations. S5. In the scalp surface coordinate system, select the scalp surface above the ear plane as the base area of ​​the head cover, expand along the outer normal vector of this area to form a three-dimensional model of the individualized positioning head cover, and reserve the coil placement area corresponding to the individualized scalp target point according to the individualized scalp target point position. S6. Based on the three-dimensional model, create an individualized positioning headgear for the repeated and standardized placement of the transcranial magnetic stimulation coil in terms of spatial position and orientation.

2. The personalized wearable transcranial magnetic stimulation therapy-assisted positioning method according to claim 1, characterized in that, In S1, the preprocessing methods for resting-state functional magnetic resonance imaging include: removing the first 10 time points, time-level correction, spatial correction, standardization, smoothing to remove linear drift, bandpass filtering, regression covariates, and cubic spline interpolation.

3. The personalized wearable transcranial magnetic stimulation therapy-assisted positioning method according to claim 1, characterized in that, S2 specifically refers to: S21. Model and segment the scalp surface, skull surface, and cerebral cortex surface based on T1-weighted structural imaging. The specific expressions for modeling and segmenting the scalp surface, skull surface, and cerebral cortex surface are as follows: In the formula, voxels Organization category label, Used for voxels Marked as At that time, the measurement of "voxels" "Actual grayscale value" and "Organization category" The degree of inconsistency between the "gray intensity models" It is a neighborhood set between voxels. For smoothing terms, voxels within the neighborhood Organization category label, To be related to voxels Forming neighborhood pairs Another voxel index, For in the marked field The total energy function under the following conditions These are the weighting coefficients for the smoothing term, used to balance the relative contributions of the data term and the spatial smoothing term; S22. Reconstruct the scalp surface, skull surface, and cerebral cortex surface based on the modeling and segmentation results. The specific expressions for reconstructing the scalp surface, skull surface, and cerebral cortex surface are as follows: In the formula, The total energy function reflects the degree of deviation between the current surface shape and the desired solution. Let be the set of three-dimensional coordinates of all vertices on the surface to be reconstructed. Energy is the data term used to measure the consistency between the reconstructed surface and the grayscale / boundary information of the original magnetic resonance image. The smoothing energy is used to measure the local smoothness of a surface. These are the weighting coefficients for the data items, used to adjust the degree of importance attached to the image data. This is the weighting coefficient for the smoothing term, used to adjust the importance of surface smoothing constraints.

4. The personalized wearable transcranial magnetic stimulation therapy-assisted positioning method according to claim 1, characterized in that, In S3, the specific expression for performing whole-brain functional connectivity analysis is: In the formula, For the first i The strength of the functional connectivity between voxels and seed sites t For time point indexing, T This represents the total number of time points in the resting-state magnetic resonance time series. For at a certain point in time t At that time, the time series mean of the BOLD signal in deep nuclei, The seed time series is the average value over the entire time period. , For at a certain point in time t At that time, the first i Time series values ​​of the BOLD signal of individual voxels For the first i The mean of the individual time series. .

5. The personalized wearable transcranial magnetic stimulation therapy-assisted positioning method according to claim 1, characterized in that, S4 includes the following steps: S41. Determine the grid vertex coordinates of anatomical landmarks on the 3D grid of the scalp. The anatomical landmarks include the root of the nose, the external occipital protuberance, the left preauricular point, and the right preauricular point. S42. Calculate the geodesic distance from any scalp vertex to the anatomical landmark using the surface geodesic distance algorithm, and obtain the geodesic distances of the root of the nose, external occipital protuberance, left preauricular point and right preauricular point; S43. Using normalization, construct the horizontal coordinate based on the difference in geodesic distance between the left and right preauricular points, and construct the vertical coordinate based on the difference in geodesic distance between the nasal root point and the external occipital protuberance, forming a scalp surface coordinate system constrained by four anatomical points. S44. In the scalp surface coordinate system, the individualized cortical target point is projected onto the scalp surface along the radial or shortest path to obtain a unique scalp target point coordinate, which serves as the individualized scalp target point position to guide coil placement.

6. The personalized wearable transcranial magnetic stimulation therapy-assisted positioning method according to claim 1, characterized in that, S5 specifically refers to: Based on the area above the ear plane determined by the line connecting the left and right preauricular points, the upper half of the scalp surface of the top of the head and the top of the forehead is selected as the base area of ​​the headgear in the scalp surface coordinate system. A thin shell structure is formed by extrusion along the outer normal vector to generate a three-dimensional model of the individualized positioning headgear that matches the patient's head shape. According to the individualized scalp target point position, a coil placement area corresponding to the individualized scalp target point is reserved on the surface of the individualized positioning headgear.

7. The personalized wearable transcranial magnetic stimulation therapy-assisted positioning method according to claim 1, characterized in that, S6 specifically refers to: The 3D model is exported as a 3D printable data format, and an individualized positioning headgear is made by 3D printing. The outer surface of the headgear retains positioning marks and strap fixation structures corresponding to the target points, thereby enabling the repeated placement and standardized operation of the transcranial magnetic stimulation coil spatial position and orientation among different operators.

Images