Method and system for displaying the electric field generated on the brain by transcranial magnetic stimulation

Method and system for displaying the electric field generated by transcranial magnetic stimulation in the brain

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] The present invention claims the benefits of U.S. Provisional Application No. 60/825,454 filed on September 13, 2006, which has been assigned to the assignee of this application and incorporated herein by reference.

Technical field

[0003] The present invention generally relates to transcranial magnetic stimulation, and more specifically, to display an accurate representation of the electric field induced by a transcranial magnetic stimulation induction coil device in the brain of a subject.

Background technique

[0004] Transcranial magnetic stimulation (TMS) uses an induction coil in which a time-varying magnetic field is generated to induce an electric field in the brain ("Ef-defined"). Neurons located in the brain that are exposed to a sufficiently strong electric field will be activated or stimulated. In navigated brain stimulation (NBS), the electric field induced by the TMS induction coil in the brain is shown as an overlap on the graphic display of the anatomical representation of the subject's brain. By observing the display, the user can visualize the electric field induced in the brain, and thereby interactively position the TMS coil device relative to the brain in real time to stimulate the target location of the brain.

[0005] The following data acquisition and processing steps are usually implemented as part of NBS.

[0006] 1. Generate a segmented data representation of the scalp or head surface of the subject based on data representing the anatomical structure of the subject's head. Generally, well-known software algorithms are used to process two-dimensional ("2D") magnetic resonance imaging ("MRI") data representing the subject's head to generate a three-dimensional ("3D") representation of the volume of the head. Among them, the data representing two-dimensional magnetic resonance imaging was previously obtained using traditional MRI technology, and the image includes at least the human brain, the upper part of the skull, and attached tissues and cartilage. Then, the 3D representation of the head is further processed by a known software algorithm to generate a segmented data representation of the head surface of the object.

[0007] 2. Implement a tracking element to track the position and direction of the TMS coil device relative to the head. Like the conventional technology in this field, easily recognizable reflection marks (trackers) are placed on the selected points of the subject's head and the TMS coil device, so that the 3D coordinates and six degrees of freedom of these points can be automatically recorded. For example, the tracker on the TMS coil device can be a part of the tracking device attached to the TMS coil device, such as the submission on August 30, 2007 entitled "TRANSCRANIAL MA GNETIC STIMULA TION IND UCTION COIL DEVICE WITH ATTACHMENT PORTION FOR RECEIVING TRACKING DEVICE" described in US Patent Application No. 11/847,544, which is assigned to the assignee of this application and incorporated herein by reference. Like the conventional technology in this field, a dedicated camera is used to record the coordinates of the tracker.

[0008] 3. Perform registration processing that combines the data (the second point above) representing the position of the TMS coil device and the tracker on the subject's head during the tracking calibration process and the image data that generates a 3D representation of the subject's head (Point 1 above) is related. Generally, multiple landmark points (for example , points on each ear or nose) with reflection trackers on the head are accurately positioned on the 2D MRI image or volumetric 3D image (if available) of the head. The same point can also be accurately positioned on the subject's head by using a digital pen tracker including a reflection tracker. After performing this point-to-point correspondence or point-to-point matching, calculate the coordinate system used to represent the head in the MRI image data and the coordinate system used to represent the tracking during the calibration process.

CN 101517618 Β

The coordinate system alignment transformation of the relative position of the detector. By performing additional point-to-point matching, for example, the quality of the transformation is enhanced at least in the sense of the least squares method, thus improving the accuracy of the NBS.

[0009] 4. On a display commonly used in NBS, a graphical representation of the TMS coil device relative to the graphical representation of the scalp and brain at the selected depth is displayed. It is particularly preferred to display only the TMS coil device containing coil windings. shell. The electric field induced by the TMS coil device in the brain is shown as an overlap on the representation of the brain. Therefore, when the user navigates the TMS coil device relative to the head of the subject, the display provides the user with a visual representation of the position and direction of the TMS coil device housing relative to the head and brain, thereby providing the coil windings relative to the head and the brain. A visual representation of the location and direction of the brain, and provides an electric field induced in the brain. The quality of the transformation calculated in the registration (point 3 above) affects the accuracy of the representation shown on the display and thus the navigation accuracy. It is well known that the electric field induced by the coil windings is calculated using a head model or a head conductivity distribution model (such as a spherical model), such as Ravazzani, The model described by P. et al. in the article Magnetic stimulation of the nervous system: induced electric field in unbounded, semi-infinite, spherical, and cylindrical media (Annals of Biomedical Engineering 24: 606-616, 1996), the entire document The content is incorporated herein by reference, and the model is based on a model of the shape and position of the copper wire winding in the TMS coil device. Then, the electric field is shown in the representation of the brain, for example, the intensity of the electric field is indicated by color, so that the user can navigate the TMS coil device to stimulate the target point of the brain. To a large extent, the accuracy of the brain representation determines the accuracy of the representation of the electric field induced in the brain shown on the display and greatly affects the accuracy with which the user navigates the TMS coil device to stimulate the target point on the brain.

[0010] It is known that the effect of TMS depends on the absolute intensity of the electric field of the target point on the brain and its relative intensity relative to the adjacent area of the target point. Therefore, it is very important for the NBS display to accurately show the user of the TMS coil device the position of the maximum electric field in the neighborhood of interest of the target point. For anatomical reasons, on existing NBS displays, the neighborhood of interest at a selected depth of the brain is usually expressed as a direction approximately parallel to the surface of the subject's scalp above the neighborhood. Therefore, in the existing NBS, multiple so-called visualization surfaces are used to represent the brain, and each visualization surface represents a part of the brain at a selected depth.

[0011] In the prior art, an NBS display usually displays a visualization surface at a depth of approximately 20 to 25 mm below the scalp. The shape of the visualization surface approximates the shape of the brain and cortical brain structure at the selected depth. In addition, the NBS display displays the electric field induced at points along the visualization surface (including the location of the largest electric field on the visualization surface). The electric field is usually shown as a color map, where the colors indicate the intensity relative to the maximum. In addition, NBS displays usually show the TMS coil device in color relative to the visualization surface. In addition, real-time updated, polygonal colored and textured surfaces are also used to display the visualized surface, for example, when the TMS coil device or the head is moved in real time. In addition, on existing NBS displays, the TMS coil device and the visualization surface can be viewed from any angle and distance.

[0012] In the existing NBS, the visualization surface is directly obtained from the 2D MRI image data of the head, so that any protrusions (protrusions), depressions, or other irregularities in the scalp are correspondingly and substantially the same reflected in the visualization surface . Ideally, it is expected that the position of the strongest electric field will move along the visualization surface on the NBS display in response to the movement of the TMS coil device along the scalp.

[0013] However, it has been observed on the existing NBS display displaying the visualization surface that when the TMS coil device is gradually moved along the subject's scalp, the electric field representation on the visualization surface becomes irregular and does not completely correspond to the TMS coil along the scalp. Movement, where the visualization surface displayed on the existing NBS display has obvious ridges, depressions, or irregularities relative to the TMS coil device and the electric field induced by the TMS coil device. For example, if the stimulus target point is located at the bottom of the surface depression of the visualization surface, the electric field intensity at the target point displayed on the visualization surface is significantly smaller than the points around the target point, even though

CN 101517618 Β

The actual electric field intensity at the target point may be the same or similar to the electric field intensity of the surrounding points. It is well known in the art that the electric field induced by the TMS coil device rapidly weakens with the distance from the coil winding in the TMS coil device, for example, the electric field at a distance of 2 mm from the coil winding can be weakened even by about 5-20%. Therefore, if the visualization surface is very close to the shape of the scalp, the maximum electric field will most likely not appear as a depression in the visualization surface that constitutes the target point of stimulation, but as a point on the visualization surface that is located near the target point and at the edge of the depression. Place. Therefore, when the visualization surface includes depressions corresponding to the target point, it will be difficult, if not impossible, to locate the TMS coil device so that the electric field is located at the target point on the visualization surface. Optionally, if the visualization surface is not very close to the brain shape, the area near the target point will probably not be oriented correctly, making it difficult to accurately position the TMS coil device relative to the head using NBS to induce induction at the target point The target of the strongest electric field.

[0014] Therefore, it is necessary to generate a visualization surface representing the part of the brain at a selected depth to display the electric field induced by the TMS coil device in the brain as part of the NBS. The visualization surface accurately represents the brain at the selected depth and avoids visualization An incorrect representation of the electric field appears on the surface.

Summary of the invention

[0015] According to the present invention, the visualization surface representing the brain part of the subject at the selected depth is achieved by comparing the actual representation of the subjects head surface obtained from the image data representing the anatomical structure of the subjects head with the ideal representation of the head surface. It is generated by the combination, wherein the combination eliminates or substantially reduces any protrusions, depressions or other irregularities on the visualization surface, and provides a TMS coil device for stimulating the subjects brain in the part of the brain corresponding to the visualization surface. The actual representation of the electric field. The contribution of each of the actual and ideal representations of the brain surface is a function of the depth of the visualization surface, and the size of the visualization surface generated by this combination is proportional to the depth of the visualization surface.

[0016] In one embodiment, the visualization surface generated by this combination is substantially convex. In another embodiment, the ideal representation of the head surface is a function of the raised area of the scalp, and there are no or substantially no protrusions, depressions or other irregularities in the scalp (which have been removed or substantially reduced). In yet another embodiment, the ideal representation of the head surface is an ellipsoid.

[0017] In yet another embodiment, the visualization surface generated by the combination is compared with the actual representation of the head surface, and if the difference between the former and the latter exceeds a predetermined threshold, the actual representation is adjusted to the ideal representation. Combine the contributions (ie, the difference function) to generate a new visualization surface. In one embodiment, the difference is an average value that defines the distance between the actual representation of the head surface and the multiple corresponding points of the visualization surface.

Description of the drawings

[0018] According to the following detailed description of the current preferred embodiments in conjunction with the accompanying drawings, other objectives and beneficial effects of the present invention will be apparent, and similar numerals in the accompanying drawings indicate similar elements.

[0019] FIG. 1A is a cross-sectional view of an exemplary head surface portion of an object, which includes actual and ideal representations corresponding to the head surface and visualization of the contour of the electric field induced by the TMS coil device located above the head surface Surface; [0020] FIG. 1B is an enlarged view of a portion of the head surface shown in FIG. 1A;

[0021] FIG. 2A is a cross-sectional view of the head of the subject of FIG. 1A, which includes a visualization surface corresponding to an actual representation of the head surface;

[0022] FIG. 2B is a cross-sectional view of the head of the subject of FIG. 1A, which includes a visualization surface corresponding to an ideal representation of the head surface;

CN 101517618 Β

[0023] FIG. 4A is an exemplary surface mesh representation of the subject's head;

[0024] FIG. 4B is an exemplary scalp mesh representation of the head of FIG. 4A according to the present invention;

[0025] FIG. 5 is a perspective view of a three-dimensional representation of an exemplary head according to the present invention, on which the visualization surface of the brain is covered, and shows the electric field induced by the TMS coil device on the visualization surface.

Detailed ways

[0026] In NBS, previously collected MRI image data representing the anatomical structure of the subject's head is used to generate a three-dimensional ("3D") image representing the head and a brain part representing the scalp or a selected depth below the surface of the head The visualization surface. The NBS display usually displays the visualization surface superimposed on the 3D image of the head, the position of the TMS coil device used to stimulate the target point on the brain relative to the head surface and the visualization surface, and the brain surrounding the target point on the visualization surface Electric field induced in the area.

[0027] In the existing NBS, the structure of the generated visualization surface basically corresponds to the actual structure of the head surface of the subject. 1A shows a typical head 10, a visualization surface 12 of the brain under the scalp 14 of the head 10 (which is generated according to the prior art), and a basic idealization of the brain part of the head 10 at the same depth as the surface 12 A cross-sectional view of the visualization surface 32 of the scalp, the TMS coil device 16 located above the scalp 14, and an isoplanar contour 18 representing the electric field induced by the TMS coil device 16 at various depths below the scalp 14. For ease of explanation, the device 16 shown in FIG. 1A is located slightly above the scalp 14, although the device 16 is usually located directly on the scalp 14 in practice. 1A, the visualization surface 12 includes bumps, protrusions, and irregularities 13, which basically correspond to the surface shape present in the scalp 14, that is, bumps, protrusions, and irregularities 15. FIG. 2A is a view of the head, which shows only the visualization surface 12. 2A, the visualization surface 12 includes a point IL that will be displayed on the visualization surface when the TMS coil device 16 moves along the scalp 14 and represents the strongest electric field induced in the brain at the depth of the visualization surface 12. In the prior art In this case, the visualization surface 12 generally does not include points 11 in any of the depressions 13 in the visualization surface 12 that correspond to the depressions 15 covered on the scalp 14.

[0028] FIG. 1B is an enlarged view of a portion of FIG. 1A, in which the scalp 14 includes depressions 15A. 1A and 1B, the electric field induced by the TMS coil device 16 has a maximum value at the point where the visualization surface (for example, the visualization surface 12) intersects the contour 18 having the maximum electric field value. In the exemplary embodiment of the TMS coil device 16 shown in FIGS. 1A and 1B, the target direction or the focus direction of the TMS coil device 16 corresponds to the line AA that extends orthogonally from the TMS coil device 16 and the line AA extends through the recess 15A. The point A in the middle and the point E in the depression 13A on the visualization surface 12, the point A is located on the contour 18C and the distance from the TMS device 16 is L1, and the point E is located on the contour 18E and the distance from the TMS device 16 is L2. In addition, for exemplary purposes only, the scalp 14 is considered to have conductivity similar to that of the brain, so that when the TMS coil device 16 is in the position shown, the strongest electric field induced in the scalp 14 is located in the recess 15A. Point A. On the existing NBS display, the strongest electric field induced by the TMS coil 16 will be represented as points D1 and D2 on the visualization surface 12. The position of the TMS coil 16 is suitable for stimulating the target point at the midpoint E of the recess 13A in the visualization surface 12. Points D1 and D2 correspond to points on the contour 18D that are separated from the line AA and are close to the recess 13A. Although the point E in the recess 13A of the visualization surface 12 is a stimulation target, due to the shape of the recess 13A, the strongest electric field represented on the visualization surface 12 will not be located at the point E in the recess 13A. Similarly, if the stimulus target point is located below the point E, the visualization surface 12 will be located at a deeper depth in the brain, so that the actual target point of the point E is no longer on the visualization surface 12. In addition, if the stimulation target point is located above point E, for example on the idealized surface 32, the actual target point of point E will also be located on the idealized surface 32. Thus, based on the maximum electric field represented on the idealized surface 32, TMS The coil device 16 also cannot be located at the target point where the stimulus is actually at point E. Therefore, due to the shape of the recess 13A, the specific

CN 101517618 Β

At the depth of the brain surface, the TMS coil device 16 cannot be positioned to point to a target point below, above, or at point E. [0029] It is well known that in NBS, it is desirable to recognize and know the position of the TMS coil device relative to the scalp, and the TMS coil device induces the strongest electric field at a target point in a certain depth of the brain. Referring to the exemplary head 10 shown in FIGS. 1A and 1B, when the target position on the brain part corresponding to the depth represented by the visualization surface 12 is aligned with the point A on the scalp 14 of the head 10, the visualization surface 12 The recesses 13A in, generate a representation of the maximum electric field, as shown by points D1 and D2 in the visualization surface 12. Therefore, because the maximum electric field is shown at points D1 and D2, the position of the TMS coil device 16 that will induce the maximum electric field at the target position E of the brain part at the selected depth relative to the scalp 14 cannot be determined. In addition, if the TMS coil device 16 is moved slightly in the direction of arrow R or L along the surface of the scalp 14, the maximum electric field shown on the visualization surface 12 will likely still be at or near D1 and D2, only at D1 Or near, or only at or near D2. Therefore, since the target position on the brain part is aligned with the depression 15A in the scalp 14, the position of the TMS coil device 16 relative to the scalp 14 cannot be accurately determined for stimulation corresponding to the scalp 14. The target location of the part of the brain directly below point A. Therefore, the position of the TMS coil device 16 cannot be determined and stored to provide repetitive, precise, and maximum stimulation to the target location on the brain aligned with the point A on the scalp 14.

[0030] Ideally, it is desirable to represent the brain part of the subject at a selected depth under the scalp by using a smooth or substantially convex curve, because this can accurately identify the TMS coil that generates the largest electric field at the target location of the brain The position of the device relative to the subject's head. 1A and 1B, if an ideal visualization surface 32 is used to represent the part of the brain at a selected depth, then the maximum electric field of the TMS coil device 16 aligned with point A on the scalp 14 will only represent the point on the visualization surface 32 At D, point D is the intersection of line AA and contour 18D and the distance from TMS coil device 16 is L3_O2B, FIG. 2B only shows a schematic view of the head 10 of the visualization surface 32. The smooth convex surface 32 allows when the TMS coil device 16 gradually moves along the scalp 14 (for example, in the direction of the arrow R or L), The point 17 representing the maximum electric field induced on the surface 32 by the TMS coil device 16 is also substantially along the surface 32 accordingly. However, the visualization surface must also be very close to the shape of the brain in order to correctly orient the area adjacent to the target location to accurately identify the position of the TMS coil device relative to the head that generates the largest electric field at the target location.

[0031] According to the present invention, a visualization surface representing the brain part of the subject at a selected depth is generated so that when the TMS coil device is gradually moved on the subject's scalp, the TMS coil device is shown relative to the visualization surface and the subject's head. On the NBS display of the anatomical position, the representation on the visualization surface of the maximum electric field induced by the TMS coil device at the target location on the brain part of the selected depth moves along the visualization surface accordingly. Therefore, the position of the TMS device with the maximum electric field induced at the target position on the brain part of the selected depth relative to the target's head can be identified, so that the TMS coil device provides repeated and accurate stimulation of the target position.

[0032] The exemplary process for generating a substantially convex visualization surface of multiple depths of the brain according to the present invention can accurately identify the TMS coil device that induces the largest electric field at the target position on the brain of the selected depth. The position of the head of the subject. It will be understood that the data processing steps in the exemplary process can be easily performed by a processor with sufficiently high processing power to provide output data on the display in real time or substantially in real time, and the processor includes a memory and is coupled to For displays such as monitors and data input devices such as mice and keyboards. For example, the processor may be a microprocessor programmed to perform TMS in conjunction with a TMS coil device, such as US Patent No. 6,849, 040 assigned to the assignee of this application and incorporated herein by reference. In order to highlight the nature of the present invention, an exemplary process is described in conjunction with the exemplary head 10 described above. It will also be understood that although the data calculation and processing of the various steps in the exemplary process may be well-known or customary in the art, the combination of the steps of the exemplary process is creative, and the resulting visualization surface has Innovative and superior to the prior art visualization surface, such as the use in NBS.

CN 101517618 Β

[0033] Using data processing techniques known in the art, the previously acquired 2D MRI image data representing the anatomical structure of the head 10 is combined and segmented to generate a volume image of the head defined by a plurality of voxels. Similarly, the coordinates of the voxels that define the surface of the head or the scalp 14 of the head 10 are recognized using known techniques, and then stored in the memory. [0034] Then, a polyhedral surface mesh 100 representing the surface of the head 10 is generated using data processing techniques known in the art, as shown in FIG. 4A, in which the surface mesh 100 is defined by a plurality of polygons 102. A typical polygon is a triangle with a length of about 1 to 2 mm and a width of about 1 to 2 mm. 4A, the size of each polygon 102 is proportional to the size of the voxel, so that the surface mesh 100 has substantially the same level of detail (resolution) as the MRI image data. For ease of reference, the surface grid 100 is hereinafter referred to as the actual representation of the head surface 14 or Rep-Actual.

[0035] Using data processing techniques known in the art, the smallest convex mesh or convex hull that includes or surrounds the surface mesh 100 is generated. In addition, data processing techniques known in the art are still used to generate a bounding box that includes the surface mesh 100 and defines a set of coordinates, where the head 10 is in the bounding box.

[0036] Still using data processing techniques known in the art, the polygons that define the lower part of the head 10 (such as the part below the ears) are identified, and then a new convex mesh is generated after the polygons are removed. In addition, we still use data processing techniques known in the art to identify polygons with areas exceeding a predetermined area in the new convex mesh ("large polygons") and polygons sharing vertices with the large polygons ("neighbor polygons"), and then pass Remove large polygons and neighboring polygons, and generate a synthetic convex mesh from the new convex mesh. The resultant convex mesh is the scalp mesh 110, as shown in FIG. 4B.

[0037] For a typical head, removing adjacent polygons from the convex mesh corresponds to removing isolated portions of the convex mesh corresponding to the raised sub-regions of the head surrounded by the recessed region. 4A and 4B, due to the geometric properties of the convex mesh and the relatively small size of the polygons 102 in the surface mesh 100, the large polygons of the convex mesh correspond to the depressions in the surface mesh 100, for example, between the ears and the skull. The recessed area between. 4B, the convex mesh includes polygons 102 that form the skull cap shape of the head 10 with the protrusions and holes removed. A few polygons 102A are located in areas where the head 10 has a bulge, and most polygons 102B are densely located on the head 10. Other areas. Still referring to FIG. 4A, it can be seen that the raised sub-region above the left ear 103 of the head 10 generates many polygons 102, which are used to adapt the ideal representation of the head surface such as an ellipse to the actual head surface of the object. It means, and the irregular sub-region 108 on the left temple 105 of the head 10 only generates a small number of polygons.

[0038] The vertices of all remaining polygons in the scalp mesh 110 are identified and stored. Based on the vertices of the remaining polygons, the estimated value of the centroid μ of the brain of the head is calculated using a known technique. Then, by using the least square method, an ideal convex shape, for example, an ellipsoid, located around the center of mass and suitable for the extracted vertices of the scalp mesh 110 is formed. The ideal convex shape is suitable for the convex area of the scalp 14, and the depression in the scalp 14 is basically ignored. In one embodiment, the ellipsoidal surface is a sufficiently ideal convex shape, because if the TMS coil device is located on or near the outer surface of an ellipsoidal body with conductive features (such as those already in the brain), and Gradually moving along the outer surface of the ellipsoidal body, the electric field induced along the outer surface of the ellipsoid will correspondingly change with the movement of the TMS coil device. For ease of reference, the generated ideal convex shape representation of the head surface is referred to as Rep-Ideal ο below.

[0039] In a preferred embodiment, a technique known in the art is used to generate the topmost bounding box of the head 10 or the skull cap of the head 10. According to the content of the bounding box, calculate the estimated value of the center of mass U of the brain. Then, an ellipsoid with a centroid of μ and suitable for the apex of the generated scalp mesh 110 is generated. As described above, the vertex density of the scalp mesh 110 is greatest at the top of the area on the skull cap corresponding to the most convex area of the actual head surface represented by the surface mesh 100. Therefore, the substantially convex area of the surface grid 100 has the greatest influence on the shape of the ellipsoid. In a preferred embodiment, the ellipsoidal surface is adapted to the surface of the head by using a well-known technique that minimizes the minimum mean square error (for example, pseudo-inverse)

CN 101517618 Β

The apex.

[0040] Combine Rep-Actual and Rep-Ideal to generate a synthesized visualization surface V (D), as shown below:

[0041] V (D) = Wactuai*Rep_Actual+Wideai*RepTdeal (1)

[0042] Wherein, Wa "ual and Wide are weighting factors, and D is the distance between the scalp 14 and the stimulation target position on the brain part of the head 10 ("the depth of the synthesized visualization surface" or "the depth of peeling") Therefore, the synthesized visualization surface is the weighted average of the surface grid 100, that is, the representation of the actual head surface and the ideal convex shape representation of the head surface generated by the scalp grid 110.

[0043] According to the present invention, by balancing the following objectives, equation (1) is used to generate a visualization surface of the brain at a specific depth with a structure that is advantageous for use in NBS.

[0044] 1. The shape of the brain part represented by the visualization surface should be anatomically reasonable. Because the volume segmentation of the brain is not performed routinely, the anatomical rationality is a function of how similar the shape of the brain is to the shape of the scalp or head surface.

2. The shape of the part of the brain represented by the visualization surface should be sufficiently smooth or convex so that the electric field represented on the visualization surface corresponds to the actual state of the electric field at the depth of the visualization surface in the brain. The second goal is clearly contradictory to the first goal. This is because if the brain is a perfect sphere, then the actual representation of the electric field on the visual surface should be ideal.

[0046] Because the objective 1 and objective 2 are contradictory according to the present invention, the weights Wa"ual and Wide in equation (1) are functions of D. In order to increase D, the influence of objective 1 is reduced, so that the weight of Rep-Actual Decrease, which leads to an increase in the weight of Rep-Ideal. Because any surface of the brain will not be visible in the deeper part of the brain, therefore, for the deeper peeling depth, apply this in formula (1) Weighting minimizes or no errors in the construction of the synthesized visualization surface. Therefore, the micro cortex that protrudes deeper in the brain does not have any effect. In contrast, when the skinning depth is close to 0, according to target 1, The weights of Rep-Ideal are the same, and the synthesized visualization surface is close to the structure of Rep-Actual. [0047] In a preferred embodiment, the weights Wactual and «ideal are determined by the linear function of the peeling depth, and the addition is performed The constraint condition is to prevent the weight from being less than 0 or greater than 1, so that Wa "ual+Wideal = 1. In another embodiment, Rep-Ideal is generated and appropriately modified based on experimental data related to the ideal convex shape representation of the head surface of another subject's head.

[0048] After combining Rep-Actual and Rep-Ideal to generate a synthesized visualization surface V (D), V (D) is adjusted proportionally according to the peeling depth D.

[0049] In a preferred embodiment, the user selects the peeling depth D and the axial depth AD. The axial depth AD determines parts of the head other than the part where the scalp mesh 110 is generated, such as the jaw and neck. As mentioned above, the peeling depth determines the parametric shape and zoom ratio of the generated visualization surface to represent the part of the brain at the depth D. The axial depth specifies the axial range [0, AD] within which the head surface 14 is transparent to display the visualization surface on the NBS display. In one embodiment, the peeling depth indicates that the percentage P is from 0% to 50%, and the axial depth AD is between about 0 mm and 250 mm. For example, if P = 20% and AD = 50mm, then the volume of the generated brain model is about 80% of the actual volume of the head, and the head surface 14 is transparent from the top of the head 10 down 50mm . Therefore, if the peeling depth D is selected so that ρ is close to the maximum value of 50%, then, referring to equation (l), w_actualClose to 0 and close to 1. In one embodiment, the ratio of P is: if the head model represents the head of an adult male, P is approximately equal to the number of millimeters. For example, on the top of the head, P equal to 20% corresponds to a distance measured from the brain to the nearest surface point on the scalp 14 of about 20 mm. [0050] In an exemplary embodiment, when the peeling depth P is between the minimum value and the maximum value, the weights Wactual and

CN 101517618 Β

Wideal is determined by a linear model, where:

[0051] ^actual =fo* scale+fi (2)

[0052] Where the scale = l-ρ, in other words, when the peeling depth ρ is equal to 0.1 (10%), the scale is equal to 0.9 (90%), so that the size of the visualization surface needs to be reduced It is 90% of the original scalp size. When Wactual+Wideal = 1,

[0053] Wideal = l-fo* scale-fl (3)

[0054] If Pmin and Pmax correspond to the minimum peeling depth and the maximum peeling depth, for example, 0% and 50%, the parameter f can be easily determined by rewriting the constraint condition of equation (1) as follows₀with:

[0055] f°* (1-Pmin) ⁺fl = 1 , and

[0056] f°* (lp_mJ +fj = 0 (5)

[0057] By substituting the values of Pmin and Pmax, equations (4) and (5) can be easily solved to obtain f0=2 and fi=T.

Then, you can use f₀And the known value of fl, using the P value selected by the user to calculate the scale, Wactual and «ideal °

[0058] After calculating Wactual and Wide, by expressing the shapes represented by Rep-Actual and Rep-Ideal as a finite set or equation of points, according to formula (1) the use rights Wactual and Wide will be the shape represented by Rep-Actual and RepTdeal Combine. In a preferred embodiment, when the equivalent equation becomes too complicated, the above-mentioned representation of the shape is a finite set of points. For example, it is difficult to accurately express the shape of the ear by using an equation.

[0059] In an exemplary embodiment, Rep-Actual is represented by a set of points, and Rep-Ideal of an ellipsoid is represented by an equation. In addition, the visualized surface synthesized according to formula (1) is represented by a set of points, where the number of points is equal to the number of points on the surface grid 100. This set of points corresponds to the vector V], where each vector contains the coordinates of the scalp point i, and the origin of the coordinate system is the centroid μ of the calculated ellipsoidal Rep-Ideal_οIf I: is a line that intersects both μ and V:, I: intersects the ellipsoid at two points. Further, if ei is the intersection point closest to V], then ei corresponds to V]. In this sense, e: is the projection of V: Rep-Ideal ± on the ellipsoidal surface. Therefore, the combination of these shapes is achieved through the combination of corresponding points (that is, scalp points and their projections). According to (Vi, eJ pair, calculate the new point 5 used to model the brain surface. The corresponding point is calculated as follows:

[0060] bi = Wactual*Vi+Wideal*ei

[0061] For each i, repeat the calculation of formula (6) to obtain a set of points corresponding to the model of the brain part (that is, the visualization surface at the skinning depth D). Because point e: is located on the ellipsoid surface of the ideal representation of the scalp, therefore, in the data processing stage, the brain model and the scalp mesh model of the brain have approximately the same size. Then, each point S is scaled, that is, moved closer to μ according to the peeling depth. Because the origin is at μ, each point bi is a bi* scale.

[0062] In another embodiment, Rep-Ideal is formed by a plurality of ideal convex shapes Rep-Idealn, each Rep-Ideal is formed in the convex area determined by the scalp mesh and suitable for the scalp mesh Near the center of mass μ of the brain. The use of multiple shapes (preferably, complex shapes) to form Rep-Ideal is advantageous and desirable for NBS, where the goal of TMS processing is to stimulate brain regions away from the cortical motor area. In this embodiment, multiple ideal representations of the head surface (for example, two ellipsoids) form Rep-Ideal, so that there are M projections proj and three weights W3, where:

[0063] $ = work: projj (dagger) * W/ (7 ), and

[0064]work;ί W) = 1 (8)

CN 101517618 Β

[0065] Weight Wj is equal to W_actual, w₂Is the weight of the first ellipsoid Wideall, W3 is the second ellipsoid W_ideal2The right. The projection projj is proji(Vj = V. And the projection proj? and proj\ are simple projections of the scalp point on the two ellipsoid surfaces, where proj₂(Vi)= e_iOAccording to Vi, W2 and W3 are set to 0 as follows. When V] is close to the cortical motor area of the brain, set W3 to 0. According to normalization, W2 = w_ideall, So that S is exactly the same, as if only one ellipsoid is used to form Rep-Ideal. In contrast, when% is far away from the cortical motor zone, set w? to 0, and W3 = w_ideall, To calculate 5 according to the second ellipsoid.

[0066] After adjusting V(D) in proportion to the peeling depth D, according to the position of the TMS coil device 16 relative to the scalp 14, calculate the visualized surface V(D) at the peeling depth selected by the user. electric field. Then, a three-dimensional image of the brain anatomy that contains the electric field induced by the TMS coil device and is covered by the visualization surface is displayed on a monitor as associated with performing NBS, as shown in FIG. 5. In a preferred embodiment, the maximum electric field at the target position and the electric field in the area near the target position are displayed on the visualization surface. Referring to FIG. 5, the electric field 120 is shown to be overlaid on the image of the brain anatomy 122 of the head 10. In the preferred implementation on the display, the colored electric field 120 is overlaid on the anatomy 122 in gray scale, thus allowing the user to easily distinguish between the two types of information. The dark spot 124 under the white cylinder 126 represents the large value of the electric field, which is preferably represented by dark red, and the dark area 125 near the edge 130 of the circumference of the electric field 120 represents the minimum value of the electric field, which is preferably represented by dark blue .

[0067] In one embodiment of the present invention, the inner surface of the skull is used instead of the inner surface of the scalp to generate

Rep-Actual, for example, when the external shape of the head does not accurately conform to the shape of the cranial cavity.

[0068] In another embodiment, Rep-Actual is generated based on the actual shape of the brain or cortex, as determined by the segmentation of the MRI image. By using this Rep-Actual, the visualization surface generated by equation (1) can have increased anatomical fidelity, while also exhibiting irregularities representing the electric field induced on the visualization surface.

[0069] In yet another embodiment, the user can select one of a plurality of additional constraint conditions to be applied to equation (1) to generate a visualization surface. These constraints, for example, correspond to and are based on experimental results obtained by generating visualization surfaces of other heads with a head surface configuration similar to that of the subject's head; these constraints are also selected to avoid generating users It will be difficult to interpret the visualization surface of the electric field information displayed on the visualization surface.

[0070] In yet another embodiment, Rep-Ideal is obtained by using one or more selected optimization criteria and constraints according to the current position of the TMS coil device. The optimization criteria and constraints may include, for example, generating a visualization surface only at a local area known in advance close to the target location of the stimulation. For example, in the MRI image data representing the head, if the data representing the right hemisphere of the brain is incomplete and the data representing the left hemisphere of the brain is complete and complete, and further, if the predetermined stimulation target position of interest is located on the left Hemisphere, then, only the part of the surface mesh 100 close to the predetermined position is used to fit the curved surface of the desired depth. In the fit of the curved surface, the part of the surface grid 100 far away from the stimulation position is ignored.

[0071] In another embodiment, a microcontroller for executing NBS and capable of generating a visualization surface in accordance with the present invention uses information representing the position of the TMS coil device relative to the head and the electric field induced on the visualization surface, To automatically control or provide the user with control information to control the position or direction of the TMS coil device relative to the head, so that the electric field induced on the visualization surface remains constant when the TMS moves along the surface of the head. The electric field with a constant amplitude can be defined by the user, for example. This controlled application of electric fields has special uses, for example, when the magnitude of the maximum electric field depends on the position or direction of the TMS coil device relative to the scalp, and when it is desired to stimulate the target area of the cortex with the TMS coil device in a uniform manner.

[0072] In yet another embodiment, the coordinates of the visualization surface are compared with the corresponding coordinates of Rep-Actual.

CN 101517618 Β

To determine whether the difference between the shape of the visualization surface and the shape of Rep-Actual is within a predetermined threshold. If it is, then V(D) is adjusted proportionally according to the peeling depth D. If no, then w_actualOr w_idealA new value is specified or calculated, preferably the value is a function of the difference, and Rep-Actual and Rep-Ideal are combined again to generate a visualization surface. In a preferred embodiment, the difference is the average distance between Rep-Actual and a plurality of corresponding coordinates of the visualization surface.

[0073] In yet another embodiment, if the ideal representation of the head surface of the head is determined to be unsuitable for the subject's head, ideal representations of other shapes (for example, ellipsoids of other shapes) may be used as Rep- Ideal ο

[0074] Although the preferred embodiments of the present invention have been described and elucidated, it will be apparent to those skilled in the art that various modifications can be made without departing from the principles of the present invention.

CN 101517618 Β Attached drawings

<img file="CN101517618B_D0001.tif" />

Figure 1Α

CN 101517618 Β