Method for determining a tumor resection path in a breast-conserving surgery planning and related products

Abstract

The application discloses a method for determining a tumor resection path in breast cancer breast-conserving surgery planning and related products. The method comprises the following steps: acquiring MRI images of a patient in a prone position transmitted from a cloud server; performing three-dimensional model reconstruction based on the MRI images in the prone position, and extracting at least a tumor model and a breast skin model on the affected side; calculating the tumor model and the breast skin model on the affected side in a state without external force; determining a resection path of the tumor based on a direction vector of a shortest distance point between the tumor model and the breast skin model on the affected side in the state without external force; determining a cut position according to an intersection between the resection path and the tumor model in the state without external force, and uploading the resection path and the cut position to the cloud server. By using the scheme, the shortest tumor resection path and the optimal cut position are automatically determined, and meanwhile, key blood vessels in the breast can be effectively avoided, so that the minimally invasive property and the operation safety of the breast-conserving surgery are improved.

Description

Technical Field

[0001] This application generally relates to the field of breast surgery planning technology. More specifically, this application relates to a method, electronic device, and computer-readable storage medium for determining the tumor resection path in breast-conserving surgery planning for breast cancer. Background Technology

[0002] With the development of surgical techniques, breast cancer removal surgery has evolved from the initial total mastectomy to the current breast-conserving mastectomy, which precisely removes the breast cancer, minimizes the amount of tissue removed, and preserves the appearance of the breast to the greatest extent possible. Compared to traditional total mastectomy, breast-conserving mastectomy has significant advantages in patient satisfaction, incision size, preservation of breast appearance, surgical risks, and surgical costs. However, the precision requirements for breast-conserving surgery are far higher than those for traditional total mastectomy.

[0003] In breast-conserving surgery, the choice of incision and resection path directly affects the extent of surgical trauma, the risk of vascular injury, and the aesthetics of postoperative scarring. Ideally, the path should be the shortest straight line from the surface of the breast to the center of the tumor, avoiding major blood vessels within the breast. However, currently, surgeons can only empirically choose incisions based on tumor location and surface palpation. When the tumor is deep or close to the chest wall, empirical incisions often result in excessively long paths or crossing areas of normal tissue, increasing trauma. Although three-dimensional reconstruction can show the relative positions of the tumor and blood vessels, there is a lack of effective methods for translating this information into specific, actionable surgical pathways.

[0004] In view of this, there is an urgent need to provide a scheme for determining the tumor resection path in the planning of breast-conserving surgery for breast cancer, so as to automatically determine the shortest distance tumor resection path and the optimal incision position, while effectively avoiding key blood vessels in the breast, thereby improving the minimally invasiveness and operational safety of breast-conserving surgery. Summary of the Invention

[0005] In order to at least address one or more of the technical problems mentioned above, this application proposes a scheme for determining the tumor resection path in breast-conserving surgery planning for breast cancer in several aspects.

[0006] In a first aspect, this application provides a method for determining a tumor resection path in breast-conserving surgery planning for breast cancer. The method is executed via a local server, and the method includes: acquiring a prone MRI image of a patient transmitted from a cloud server; reconstructing a three-dimensional model based on the prone MRI image, extracting at least a skin model and a tumor model of the affected breast; calculating the corresponding skin model and tumor model of the affected breast under no-force conditions based on the skin model and the tumor model; determining a tumor resection path based on the direction vector of the shortest distance point between the tumor model and the skin model of the affected breast under no-force conditions; determining the incision location based on the interface between the resection path and the tumor model under no-force conditions; and uploading the resection path and the incision location to the cloud server.

[0007] In some embodiments, the three-dimensional model reconstruction based on the prone MRI image, at least extracting the skin model and tumor model of the affected breast, includes: extracting a point set of the affected breast region from the prone MRI image based on preset MRI values ​​and target boundaries; filtering and extracting a tumor point set, a blood vessel point set, and a mammary gland point set based on the point set of the affected breast region using different MRI values, and performing three-dimensional reconstruction using triangulation to at least extract the skin model and the tumor model of the affected breast.

[0008] In some embodiments, calculating the corresponding affected breast skin model and tumor model under no external force state based on the affected breast skin model and the tumor model includes: setting material parameters for the affected breast skin model and the tumor model; superimposing gravitational potential energy into the strain energy function of the material parameters of each model to construct a total strain energy function including elastic strain energy and gravitational potential energy; calculating the displacement field of the affected breast skin and tumor tissue from a state with gravity to a state without external force based on the total strain energy function; and using the displacement field to correct the initial geometry of each model to obtain the corresponding affected breast skin model and tumor model under no external force state.

[0009] In some embodiments, calculating the displacement field of the affected breast skin and tumor tissue from a state with gravity to a state without external force based on the total strain energy function includes: deriving a momentum balance equation including the contribution of gravity from the total strain energy function, and transforming the momentum balance equation into an equivalent weak integral form; dividing the affected breast skin and tumor tissue domain into a finite number of elements in the equivalent weak integral form, and constructing a displacement shape function on each element; obtaining a set of nonlinear algebraic equations in the affected breast skin and tumor tissue domain where the global nodal internal forces and global nodal external forces are equal between each element; and iteratively solving the set of nonlinear algebraic equations to obtain the displacement field of the affected breast skin and tumor tissue from a state with gravity to a state without external force.

[0010] In some embodiments, determining the tumor resection path based on the direction vector of the shortest distance point between the tumor model and the affected breast skin model under no external force includes: obtaining the resection direction based on the direction vector of the shortest distance point between the tumor model and the affected breast skin model under no external force; and tracing the resection direction in reverse to the surface of the affected breast skin model under no external force to obtain a continuous set of surface entry points to determine the resection path.

[0011] In some embodiments, obtaining the resection direction based on the direction vector of the shortest distance point between the tumor model and the affected breast skin model under no external force includes: calculating the shortest distance point from the geometric center point of the tumor model under no external force to the surface of the affected breast skin model under no external force; and obtaining the resection direction based on the direction vector from the geometric center point to the shortest distance point.

[0012] In some embodiments, determining the resection path involves: tracing the tumor model in the absence of external force along the resection direction to obtain a continuous set of surface entry points to the surface of the affected breast skin model in the absence of external force; calculating the vertical projection of the tumor model in the absence of external force onto the resection direction based on the boundary points of the vertical projection; tracing the tumor model in the absence of external force along the resection direction to obtain a continuous set of surface entry points; and determining the resection path based on the polygonal channel formed by the set of surface entry points.

[0013] In some embodiments, the method further includes: in response to the intersection of the resection path and the vascular model or the minimum distance between the target point on the resection path and the vascular model being less than a distance threshold, calculating a first intersection volume of the resection path and the vascular model after expanding outward to the target region; performing differential angle rotation on the surface point set of the resection path in multiple orthogonal planes to obtain the corresponding rotated resection path; calculating a second intersection volume of each rotated resection path and the vascular model after expanding outward to the target region; selecting the rotated resection path corresponding to the largest difference between the corresponding second intersection volume and the first intersection volume; performing differential angle rotation on the rotated resection path corresponding to the largest difference and calculating a new intersection volume and a new rotated resection path, until the intersection volume of the new rotated resection path and the vascular model after expanding outward to the target region is zero, thereby obtaining the final resection path.

[0014] In a second aspect, this application provides an electronic device comprising: a processor; and a memory storing computer instructions for determining a tumor resection path in breast-conserving surgery planning for breast cancer, wherein when the computer instructions are executed by the processor, they cause the implementation of several embodiments of the first aspect described above.

[0015] In a third aspect, this application provides a computer-readable storage medium having stored thereon computer program instructions for determining a tumor resection path in breast-conserving surgery planning for breast cancer, wherein the computer program instructions, when executed by one or more processors, cause the implementation of the various embodiments of the first aspect described above.

[0016] Using the above-described scheme for determining the tumor resection path in breast-conserving surgery planning for breast cancer, this embodiment acquires prone MRI images of the patient and extracts a skin model and a tumor model of the affected breast through three-dimensional reconstruction. First, the corresponding models of the two models under no-force conditions are calculated, eliminating the influence of gravity on breast shape and tumor location during prone scanning, providing an accurate geometric benchmark for path planning. The tumor resection path is determined based on the direction vector of the shortest distance point between the tumor model and the skin model under no-force conditions. This path is the shortest straight line from the skin surface to the tumor, significantly shortening the surgical incision length and reducing trauma to normal tissue. Furthermore, the incision position is determined based on the interface between the resection path and the tumor model, ensuring the incision is directly opposite the tumor center, facilitating intraoperative manipulation and complete tumor removal. Thus, the shortest tumor resection path and optimal incision position are automatically determined, while effectively avoiding key blood vessels within the breast, improving the minimally invasive nature and operational safety of breast-conserving surgery. Attached Figure Description

[0017] The above and other objects, features, and advantages of exemplary embodiments of this application will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of this application are illustrated by way of example and not limitation, and the same or corresponding reference numerals denote the same or corresponding parts, wherein:

[0018] Figure 1 This is an exemplary flowchart illustrating a method 100 for determining a tumor resection path in breast-conserving surgery planning for breast cancer according to an embodiment of this application; Figure 2 This is an exemplary schematic diagram illustrating the reconstruction of a three-dimensional model according to an embodiment of this application; Figure 3 This is an exemplary schematic diagram showing the initial prone position model and the model under no external force state according to the embodiments of this application; Figure 4 This is an exemplary schematic diagram illustrating the cut path according to an embodiment of this application; Figure 5This is an exemplary structural block diagram illustrating a local server 500 for determining a tumor resection path in breast-conserving surgery planning for breast cancer according to an embodiment of this application; Figure 6 This is an exemplary schematic diagram illustrating a system 600 for determining a tumor resection path in breast-conserving surgery planning for breast cancer according to an embodiment of this application; Figure 7 This is an exemplary structural block diagram illustrating an electronic device 700 according to an embodiment of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] It should be understood that the terms "comprising" and "including" used in the specification and claims of this application indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0021] It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application. As used in this specification and claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this specification and claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations.

[0022] As used in this specification and claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if [described condition or event] is detected" may be interpreted, depending on the context, as "once determined," "in response to determination," "once [described condition or event] is detected," or "in response to detection of [described condition or event]."

[0023] The specific embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0024] Figure 1This is an exemplary flowchart illustrating a method 100 for determining a tumor resection path in breast-conserving surgery planning according to an embodiment of this application. In one embodiment, the method may be executed via a local server. Figure 1 As shown, the method includes steps S101 to S105.

[0025] First, in step S101, the patient's prone MRI image is acquired from the cloud server. Specifically, the doctor can upload the patient's original prone breast MRI file (usually in DICOM format) to the cloud server via a web browser. The cloud server then transmits this data to a local server used for performing complex computational tasks. The local server receives this image data as input for the entire planning process. In the prone position, the breast naturally droops, and the tissue distribution is closer to its natural state, facilitating subsequent biomechanical analysis and surgical planning.

[0026] Next, in step S102, a three-dimensional model is reconstructed based on the prone MRI images, and at least the skin model and tumor model of the affected breast are extracted.

[0027] In some embodiments, a point set of the affected breast region can be extracted from prone MRI images based on preset MRI values ​​and target boundaries. Based on the point set of the healthy breast region, a three-dimensional reconstruction using triangulation is performed to form a skin model of the healthy breast. Based on the point set of the affected breast region, different MRI values ​​are set to filter and extract tumor point sets, blood vessel point sets, and mammary gland point sets, and a three-dimensional reconstruction using triangulation is performed to extract at least the skin model and tumor model of the affected breast. In some implementation scenarios, a blood vessel model, a mammary gland model, and adipose tissue model may also be included to obtain a complete breast model. The aforementioned adipose tissue model can be obtained by subtracting the tumor model, blood vessel model, and mammary gland model from the skin model of the affected breast.

[0028] Specifically, the skin point set of the entire MRI scan area can first be extracted based on preset MRI values. (This typically includes the entire thoracic cavity area). It's understood that MRI values ​​(or HU values) reflect the degree to which different tissues absorb X-rays; skin, fat, glands, tumors, and other tissues have different MRI value ranges. By setting an appropriate MRI threshold, the point set of the corresponding human tissue can be extracted from the original image.

[0029] Next, the target boundary, i.e., a three-dimensional rectangle, is extracted by setting the x, y, and z axes. Based on the aforementioned target boundary, the skin point set of the entire region is... The affected and healthy breast regions were segmented and extracted, and denoted as the point sets of the affected breast region, respectively. and the healthy side breast area point set .

[0030] Point set for the affected breast region Different MRI values ​​can be set for filtering to extract tumor point sets. , blood vessel point set and mammary gland point set Next, focus on the affected breast area. Tumor point set , blood vessel point set and mammary gland point set Triangulation algorithms were applied to create skin models of the affected breast. Tumor models vascular model and mammary gland model In the implementation scenario, the data structure of each of the aforementioned models can be denoted as follows: ,in represents a point set, The vertex indices of the triangular faces. In other words, the data structures of these models all contain information about the point set and the arrangement of the triangular facet mesh.

[0031] After obtaining the above-mentioned skin model and tumor model of the affected breast, in step S103, the corresponding skin model and tumor model of the affected breast under the state of no external force are calculated based on the skin model and tumor model of the affected breast.

[0032] In some embodiments, before calculating the corresponding affected breast skin model and tumor model under no external force based on the affected breast skin model and tumor model, the method may further include: performing coordinate system transformation on the affected breast skin model and tumor model.

[0033] Because the coordinate system of the original prone MRI image is based on the scanning device, its orientation may not be consistent with the standard anatomical coordinate system. In this embodiment, the patient is in a standard prone position during breast MRI imaging, with the Z-axis typically pointing in the direction of the spine, while the X-axis (left-right) and Y-axis (front-back) orientations may vary depending on the device or scanning protocol. To ensure standardization in subsequent calculations, a coordinate system transformation is necessary. Specifically, the orientations of the X-axis and Y-axis of the reconstructed model are checked for correctness, i.e., whether the positive X-axis direction is from left to right and the positive Y-axis direction is from back to front. If the orientations are incorrect, the coordinate system can be transformed using a column exchange matrix.

[0034] In one implementation scenario, the above coordinate transformation can be expressed as: ,in The coordinate matrix of the new point set after transformation. The coordinate matrix of the original point set before transformation. The x and y axis coordinate columns are swapped. After coordinate system transformation, the point set information stored in all model files is updated to obtain the updated affected breast skin model and tumor model, and then the corresponding affected breast skin model and tumor model under no external force conditions are obtained.

[0035] In some embodiments, material parameters can be set for the affected breast skin model and tumor model; gravitational potential energy can be superimposed onto the strain energy function of the material parameters of each model to construct a total strain energy function that includes elastic strain energy and gravitational potential energy; the displacement field of the affected breast skin and tumor tissue from a state with gravity to a state without external force can be calculated based on the total strain energy function; and the initial geometry of each model can be corrected using the displacement field to obtain the corresponding affected breast skin model and tumor model in the state without external force.

[0036] For the aforementioned displacement field, in some embodiments, a momentum balance equation including the contribution of gravity can be derived from the total strain energy function, and the momentum balance equation can be transformed into an equivalent weak integral form. In the equivalent weak integral form, the affected breast skin and tumor tissue domain is divided into a finite number of elements, and a displacement shape function is constructed on each element. Based on the equivalent weak integral form and the displacement shape function, a set of nonlinear algebraic equations is obtained in which the global nodal internal forces and global nodal external forces are equal among the elements in the affected breast skin and tumor tissue domain. The nonlinear algebraic equations are iteratively solved to obtain the displacement field of the affected breast skin and tumor tissue from a state with gravity to a state without external forces.

[0037] In some implementation scenarios, the material properties of the affected breast skin model and the tumor model can be defined using a hyperelastic model. Among these, density can be defined before setting the material properties. The value ranges from 900 kg / m³ to 1060 kg / m³, or the average of the mixture is taken as 1000 kg / m³.

[0038] For tumor models Define the material properties for finite element analysis The strain energy function can be obtained based on the O'Hagan and Samani third-order Ogden model, the O'Hagan and Samani second-order polynomial model, or the first-order Ogden model. Taking the first-order Ogden model as an example, its strain energy function can be expressed as:

[0039] Where N=1, Elongation is the partial elongation. Main elongation, and For material constants, The elastic volume ratio, This is a compression parameter. As an example, it can be defined separately. , , Similarly, based on the aforementioned tumor model... Define material properties and update the tumor model. .

[0040] For mammary gland model and adipose tissue model This can be homogenized by using the same hyperelastic material parameters. Material parameters can be set using, for example, the Dempsey third-order Ogden model, a second-order polynomial model, or a second-order Mooney-Rivlin model, and the corresponding strain energy function can be obtained.

[0041] Taking the Dempsey third-order Ogden model as an example, its strain energy function can be expressed as follows:

[0042] in, Let be the strain energy function. For partial elongation, The elastic volume ratio, and A coefficient describing the shear behavior of a material. To introduce material constants for compressibility. As an example, a breast gland model can be defined separately. and adipose tissue model First-order shear modulus First-order dimensionless stiffness exponent The second-order shear modulus The second-order dimensionless stiffness exponent The third-order shear modulus The third-order dimensionless stiffness exponent Compressibility parameters The units for the aforementioned parameters are all kPa (kilopascals).

[0043] Therefore, based on the above mammary gland model and adipose tissue model Their respective material properties and This leads to an update of the mammary gland model. and adipose tissue model .

[0044] For vascular models Material properties can be defined for finite element analysis. The strain energy function, obtained from, for example, the HGO model used in vascular mechanics simulation, is expressed as follows:

[0045] Among them, matrix shear modulus Fiber modulus Incompressibility constant The remaining parameters can be referenced from the strain energy function described above. Accordingly, based on the aforementioned vascular model... Material Model Definition, updating vascular model .

[0046] In some implementation scenarios, the material settings for all the above models can be simplified to simple hyperelastic materials, i.e., only Young's modulus and Poisson's ratio are set. Alternatively, the parameter settings of the skin model can be universally applied to each sub-model, forming the same material for subsequent simulation calculations to accelerate the calculation speed. Through the definition of the above layered material models, the affected breast skin model can be obtained. Material properties Its definition Therefore, the skin model of the affected breast was updated. .

[0047] After setting the material parameters, to calculate the model under no external force, gravitational potential energy needs to be incorporated into the strain energy function. This involves applying an external force equal in magnitude and opposite in direction to gravity to restore the model to its unforced state. In some implementation scenarios, this can be achieved through... Represents gravitational potential energy, where This is the initial position of the matter point in the model. This is its current location. This corresponds to the displacement. It is the mass density at that point, because at this time gravity is recovered, specifically the gravitational acceleration. The vector form in three-dimensional coordinates is Gravitational potential energy Introducing the total strain energy function , can be defined as .

[0048] Taking the aforementioned third-order Ogden as an example, the total strain energy function, including gravitational potential energy, is:

[0049] Furthermore, based on this total strain energy function, the momentum balance equation including the contribution of gravity is first derived from the total strain energy function. Specifically, the stress formula including both elastic and gravitational components is first obtained, where the stress... Green-Lagrange strain from strain energy function The partial derivatives are obtained by calculation, i.e. For the Ogden model, the elastic stress component is... ,in For the right Cauchy-Green variable tensor, Let be the deformation gradient. For the gravitational stress component, the gravitational contribution stress can be derived using the chain rule. g.

[0050] Next, through Cauchy stress After simplifying the gravity part, we get Substituting into the momentum balance equation ,get Furthermore, the momentum balance equation can be transformed into an equivalent weak integral form by using, for example, the principle of virtual work. V= S0 indicates that the contribution of gravity to the internal forces is exactly equal to the contribution of gravity to the external forces, and the equilibrium equation degenerates into the equilibrium of purely elastic stress and surface forces.

[0051] After obtaining the weak form, the affected breast skin and tumor tissue domain is divided into a finite number of elements on the equivalent integral weak form, and a displacement shape function is constructed on each element. Specifically, a 10-node quadratic tetrahedral element (C3D10) is used to divide the continuous tissue domain. Divided into Each unit. have Each node. Within the element, the displacement field... Using nodal displacements Sum and form functions approximate: Meanwhile, the strain-displacement relationship can be defined as... ,in This is the strain-displacement matrix. Based on the weak form and displacement shape function, the internal force vector and external force vector of the element can be constructed. The element internal force vector is... (Gravity has been recovered), the external force vector of the unit is By assembling all the units, a global system of nonlinear algebraic equations can be obtained: ,in It is a global vector composed of all nodal displacements. This equation indicates that the global nodal internal forces are equal to the global nodal external forces.

[0052] Based on the aforementioned system of nonlinear algebraic equations, the displacement field of the affected breast skin and tumor tissue from a state under gravity to a state without external force is obtained through iterative solutions using methods such as the Newton-Raphson iterative method. Specifically, in the... In the next iteration, the equation will be at the current estimate. Performing a first-order Taylor expansion at point K yields the linearized system of equations: K T (U(k ))ΔU( k +1)=F ext F int (U( k )).in, This is the global tangent stiffness matrix. Solving this system of linear equations yields the displacement increments. Then update the displacement. Repeat this process until the residual is found. Less than a preset very small threshold (e.g.) When the iteration stops, the nodal displacements obtained after convergence are calculated. This is the displacement field we seek, transitioning from a state with gravity to a state without external forces. Then, this displacement field is used to correct the initial geometry of the affected breast skin model and tumor model; that is, the deformed node coordinates are... This allows us to obtain skin and tumor models of the affected breast under no-force conditions. Similarly, we can obtain models of other breasts under no-force conditions.

[0053] For example, a skin model of the affected breast under no external force conditions. Tumor models vascular model Breast gland model Fat model and the skin model of the healthy breast ,in Set parameters (including node set and element set) for the finite element mesh of each model.

[0054] After obtaining the skin model and tumor model of the affected breast under no external force conditions, in step S104, the resection path of the tumor is determined based on the direction vector of the shortest distance point between the tumor model and the skin model of the affected breast under no external force conditions. Further, in step S105, the incision location is determined according to the interface between the resection path and the tumor model under no external force conditions, and the resection path and the incision location are uploaded to the cloud server.

[0055] In some embodiments, the resection direction can be obtained based on the direction vector of the shortest distance point between the tumor model and the affected breast skin model under no external force conditions; the resection direction is traced back to the surface of the affected breast skin model under no external force conditions to obtain a continuous set of surface entry points to determine the resection path.

[0056] In some embodiments, obtaining the resection direction based on the direction vector of the shortest distance point between the tumor model and the affected breast skin model under no external force includes: calculating the shortest distance point from the geometric center point of the tumor model under no external force to the surface of the affected breast skin model under no external force; and obtaining the resection direction based on the direction vector from the geometric center point to the shortest distance point.

[0057] In some embodiments, the method of determining the resection path by tracing the resection direction backward to the surface of the affected breast skin model in a state without external force to obtain a continuous set of surface entry points includes: calculating the vertical projection of the tumor model in a state without external force perpendicular to the resection direction; and tracing the boundary points of the vertical projection backward along the resection direction to the surface of the affected breast skin model in a state without external force to obtain a continuous set of surface entry points to determine the resection path.

[0058] Specifically, from the three-dimensional reconstructed tumor model under no external force state. geometric center point Starting from the point of view, traverse the skin model of the affected breast. Calculate the nearest surface point from all points. Then the vector This is the resection direction vector. Next, the tumor model is determined along the perpendicular plane of the resection direction vector. The perpendicular projection is obtained by starting from the endpoints of the perpendicular projection polygon and following the cutting direction vector. The surface point set is then found using the direction vector surface search method. Surface point set The resulting polygonal channels form the resection path, and these polygonal channels correspond to the tumor model. The intersection of the two surfaces is the location of the cut.

[0059] In some embodiments, the method may further include: in response to the intersection of the resection path and the vascular model or the minimum distance between the target point on the resection path and the vascular model being less than a distance threshold, calculating the first intersection volume of the resection path and the vascular model after expanding outward to the target region; performing differential angle rotation on the surface point set of the resection path in multiple orthogonal planes to obtain the corresponding rotated resection path; calculating the second intersection volume of each rotated resection path and the vascular model after expanding outward to the target region; selecting the rotated resection path corresponding to the largest difference between the corresponding second intersection volume and the first intersection volume; performing differential angle rotation on the rotated resection path corresponding to the largest difference and calculating the new intersection volume and the new rotated resection path, until the intersection volume of the new rotated resection path and the vascular model after expanding outward to the target region is zero, thereby obtaining the final resection path.

[0060] Specifically, when the resection path obtained above is consistent with the vascular model When there is an intersection or the minimum distance between the target point on the resection path and the blood vessel model is less than a distance threshold (e.g., 5 mm), the surface point set of the resection path is first determined. With vascular model The set of intersection points for expanding the target area outward (e.g., expanding the area by 5mm). Simultaneously calculate the volume of the first intersection formed by the intersection point sets. Next, by using the differential rotation matrix... The surface point set of the cutting path is respectively located in the xy, yz, and xz planes. Perform differential rotation The angle is used to obtain the rotated cutting path corresponding to each plane, and its surface point set is... Furthermore, the set of surface points of the rotated cut path corresponding to each plane is calculated. With vascular model Expand the intersection set of the target region outward Meanwhile, their respective second intersection volumes .

[0061] The volume formed by the set of intersection points obtained from rotating each plane. Point set and initial volume The difference is calculated. The resection path corresponding to the fastest volume reduction (i.e., the largest difference) is selected, and then rotated again by a differential angle on its rotation plane. The new intersection volume and the new rotated resection path are calculated until the intersection volume between the new rotated resection path and the target region expanded outward from the vascular model is zero, thus obtaining the final resection path. Based on this, it is possible to avoid cutting into blood vessels while removing the tumor.

[0062] In some implementation scenarios, the above differential rotation matrix On the xy, yz, and xz planes respectively , , The above differential rotation The angle can be, for example, 0.1° or 1°.

[0063] This data is uploaded back to a cloud server, allowing doctors to view, confirm, and download it via a web interface. Based on this quantitative planning report, doctors can select the optimal surgical plan before surgery and demonstrate the expected postoperative results to the patient, greatly enhancing the certainty and safety of the procedure.

[0064] As described above, this application's embodiments ensure the biological accuracy of subsequent distance calculations by reconstructing the three-dimensional structure of prone MRI images and restoring the spatial position of tissues under no-force conditions. Based on this, the resection path is determined using the direction vector of the shortest distance point, employing deterministic calculations based on spatial geometry to eliminate path length discrepancies caused by differences in experience among different surgeons. Simultaneously, the incision location is naturally defined by the interface between the resection path and the tumor model, ensuring that the skin entry point, subcutaneous tunnel, and tumor exposure surface are collinear, avoiding additional tissue separation due to incision deviation. Data uploaded to the cloud provides direct input for subsequent intraoperative navigation equipment, establishing a digital link from image planning to surgical execution. Therefore, this application not only solves the problem of lacking quantitative basis for breast-conserving surgery incision positioning but also provides a reusable technical solution for improving the minimally invasive nature of surgery and standardizing remote collaboration.

[0065] Based on the determined tumor resection path and incision location, the surgeon can then make an incision along the path in the surgical procedure, directly reaching and completely removing the tumor in the shortest direction, while avoiding unnecessary tissue damage caused by path deviation. Furthermore, the resection path can also serve as a condition for subsequent finite element suturing simulation, used to evaluate the impact of different incision directions and cavity closure on breast appearance deformation, thus achieving closed-loop optimization from path planning to suturing outcome prediction preoperatively.

[0066] Figure 2 This is an exemplary schematic diagram illustrating the reconstruction of a three-dimensional model according to an embodiment of this application. Figure 2 As shown in the diagram, the left image is an exemplary 3D point cloud reconstruction; the right image is an exemplary 3D mesh reconstruction. As previously described, a set of points representing the region of interest is extracted from prone MRI images using preset MRI values ​​and target boundaries. This set is then merged and colored to obtain a 3D point cloud reconstruction, which shows the affected side of the breast, including blood vessels, irregular tumors, and glandular tissue. A 3D mesh reconstruction is then performed using, for example, triangulation to obtain the 3D mesh reconstruction.

[0067] Figure 3 This is an exemplary schematic diagram showing the initial prone position model and the model under a state without external force according to embodiments of this application. Figure 3 As shown in the diagram, the left image exemplifies the initial prone position model, and the right image exemplifies the model under no external force conditions, including the affected breast skin model and tumor model. For more details on obtaining the model under no external force conditions, please refer to the above. Figure 1 The descriptions in the document are not repeated here.

[0068] Figure 4 This is an exemplary schematic diagram illustrating the cut path according to an embodiment of this application. For example... Figure 6As shown in the image, the multiple cylinders indicated by the arrows correspond to the surface point set mentioned earlier. Surface point set The resulting polygonal channels form the resection path. Furthermore, the polygonal channels are related to the tumor model. The intersection of the two surfaces is the location of the cut.

[0069] Figure 5 This is an exemplary structural block diagram illustrating a local server 500 used for determining tumor resection paths in breast-conserving surgery planning according to an embodiment of this application. Figure 5 As shown, the local server 500 may include an image acquisition module 501, a model reconstruction module 502, and a computational planning module 503. In one embodiment, the aforementioned image acquisition module 501 can be used to acquire prone MRI images of a patient transmitted from a cloud server. As previously mentioned, the prone MRI images of a patient transmitted from a cloud server can be uploaded to the cloud server by a doctor via, for example, a web interface, and then transmitted from the cloud server to the local server for acquisition by the image acquisition module 501.

[0070] In one embodiment, the model reconstruction module 502 is used to perform three-dimensional model reconstruction based on prone MRI images, at least extracting the skin model and tumor model of the affected breast.

[0071] In one embodiment, the calculation and planning module 503 is used to calculate the corresponding affected breast skin model and tumor model under no external force conditions based on the affected breast skin model and the tumor model; determine the tumor resection path based on the direction vector of the shortest distance point between the tumor model and the affected breast skin model under no external force conditions; determine the incision position according to the interface between the resection path and the tumor model under no external force conditions; and upload the resection path and the incision position to the cloud server. The operations implemented by each module are the same as described above. Figure 1 The method shown corresponds to the one described above. For more details, please refer to the above. Figure 1 The descriptions made will not be repeated here.

[0072] In some embodiments, the local server 500 of this application may further include a storage module for storing the aforementioned prone MRI images and various surgical planning data, etc.

[0073] In one embodiment, this application also provides a system for determining the tumor resection path in breast-conserving surgery planning for breast cancer. This system may include a local server and a cloud server, as described in this application. The cloud server can be used to transmit prone MRI images to the local server and receive the resection path and incision location uploaded by the local server. In one implementation scenario, the system may further include a web interface, which can be used to receive prone MRI images uploaded by the doctor and transmit the prone MRI images to the cloud server, as well as receive the resection path and incision location transmitted by the cloud server. It is understood that the aforementioned web interface is also the doctor's interface. This web interface may also include, for example, a data upload module and a result review module. The doctor's interface uses this module to upload case-related data and the result review module to confirm the resection path and incision location.

[0074] In another implementation scenario, the system of this application embodiment may further include a local terminal, which can be used to receive target operations set by the surgical planning engineer for the computational planning module in a local server and target parameters adjusted for the computational planning module, and to transmit the target operations and target parameters to the local server. In some embodiments, the local terminal may further include PC-based surgical planning application software and web-based surgical planning application software, so that the surgical planning engineer can set target operations and adjust target parameters through the aforementioned software. As an example, the surgical planning engineer can set a 3D reconstruction operation and adjust the parameters of the surface point set filter in the 3D reconstruction operation through the local terminal.

[0075] Figure 6 This is an exemplary schematic diagram illustrating a system 600 for determining the tumor resection path in breast-conserving surgery planning according to an embodiment of this application. Figure 6As shown, the system 600 may include a local server 500, a cloud server 601, a web interface 602, and a local terminal 603. The local server 500 may include an image acquisition module 501, a model reconstruction module 502, a computational planning module 503, and a storage module 604. The cloud server 601 may include a cloud exchange module 605, the web interface 602 may include a data upload module 606 and a result review module 607, and the local terminal 603 may include PC-based preoperative planning application software 608 and web-based preoperative planning application software 609. During surgical planning, the doctor can upload prone MRI images or relevant medical record information to the cloud server 601 via the data upload module 606. The cloud server 601 then transmits the prone MRI images or relevant medical record information to the local server 500 via the cloud exchange module 605. Next, the image acquisition module 501 receives the prone MRI images or relevant medical record information, and the computational planning module 503 calculates the surgical planning parameters related to breast-conserving surgery for breast cancer.

[0076] The diagram further illustrates that, on the local end 603, the surgical planning engineer can set the target operations and adjust the target parameters in the calculation planning module 503 via the PC-based preoperative planning application software 608 and the web-based preoperative planning application software 609. Furthermore, relevant data from the surgical planning results can be stored through the storage module 604 and ultimately uploaded to the cloud server 601 by the local server 500. The physician can obtain the surgical planning results from the cloud server 601 on the web page 602 and confirm the resection path and incision location through the result review module 607.

[0077] As described above, this application embodiment constructs a complete set of full-chain algorithm packages for breast-conserving surgery planning for breast cancer through medical-engineering integration technology, realizing the intelligentization and automation of all steps of preoperative surgical planning, ensuring that the accuracy of all calculations is controllable and meets clinical standard requirements, while also significantly improving the efficiency of surgical planning and greatly reducing labor costs.

[0078] Furthermore, the system for determining the tumor resection path in breast-conserving surgery planning for breast cancer, as described in this application embodiment, employs a cloud + local distributed system architecture, a collaborative terminal design, and compatibility with scalable computing clusters. For example, the surgical planning system comprises a web-based data upload module, a web-based result review module, a cloud-based data exchange module, a local server computing and planning module, a local server storage module, a local PC application software module, and a local web application software module. In this system, doctors can upload images and key case data to the cloud with a single click, automatically complete preoperative surgical planning through local PC software, distribute complex calculations and data storage tasks to the local server, and return the surgical planning results to the doctor for confirmation via the cloud. Based on this, doctors do not need to learn industrial-grade software or any professional engineering skills; they only need simple training to use the system of this application embodiment to complete remote and efficient preoperative surgical planning, greatly reducing the learning difficulty for doctors in preoperative surgical planning. Simultaneously, while minimizing system costs, it further improves the overall efficiency of multi-party collaboration in surgical planning.

[0079] Furthermore, the system in this embodiment also considers the collaborative operation needs of multiple doctors and engineers, the multiple stages of requirement feedback and reconfirmation involved in the entire surgical planning process, and the multi-terminal design of the system to support real-time display and data sharing across multiple terminals. Additionally, since preoperative surgical planning emphasizes timeliness and requires doctors to review key data in real time, the system in this embodiment also includes a result review module to quickly complete the conversion of key data and provide initial feedback. It also supports multiple doctors performing multi-process tasks simultaneously, and the system's structural design supports flexible expansion of the computing module.

[0080] Figure 7 This is an exemplary structural block diagram illustrating an electronic device 700 according to an embodiment of this application. For example... Figure 7 As shown, the electronic device 700 of this application may include a processor 701 and a memory 702, wherein the processor 701 and the memory 702 communicate via a bus. The memory 702 stores program instructions for determining the tumor resection path in breast-conserving surgery planning for breast cancer. When the program instructions are executed by the processor 701, they cause the implementation of the procedure described above in conjunction with the appendix. Figure 1 The method described is for determining the tumor resection path in breast-conserving surgery planning for breast cancer.

[0081] Based on the foregoing description in conjunction with the accompanying drawings, those skilled in the art will understand that the embodiments of this application can also be implemented by software programs. Therefore, this application also provides a computer-readable storage medium. This computer-readable storage medium stores computer-readable instructions thereon for determining the tumor resection path in breast-conserving surgery planning for breast cancer. When executed by one or more processors, these computer-readable instructions implement the embodiments of this application in conjunction with the accompanying drawings. Figure 1 The method described is for determining the tumor resection path in breast-conserving surgery planning for breast cancer.

[0082] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0083] It should be noted that although the operations of the method of this application are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. On the contrary, the steps depicted in the flowchart can be performed in a different order. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0084] It should be understood that when the terms "first," "second," "third," and "fourth," etc., are used in the claims, specification, and drawings of this application, they are used only to distinguish different objects and not to describe a specific order. The terms "comprising" and "including" as used in the specification and claims of this application indicate the presence of the described features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof.

[0085] It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application. As used in this specification and claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this specification and claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations.

[0086] Although the embodiments of this application are described above, the content is merely an example adopted for the purpose of facilitating understanding of this application and is not intended to limit the scope and application scenarios of this application. Any person skilled in the art described in this application may make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in this application, but the scope of patent protection of this application shall still be determined by the scope defined in the appended claims.

Claims

1. A method for determining a tumor resection path in breast-conserving surgery planning for breast cancer, the method being executed via a local server, wherein the method includes: Acquire prone MRI images of the patient transmitted from a cloud server; Based on the prone MRI images, a three-dimensional model was reconstructed, and at least the skin model and tumor model of the affected breast were extracted. Calculate the corresponding skin model and tumor model of the affected breast under the condition of no external force based on the skin model of the affected breast and the tumor model of the affected breast; The tumor resection path is determined based on the direction vector of the shortest distance point between the tumor model and the skin model of the affected breast under no external force conditions. The incision location is determined based on the interface between the resection path and the tumor model under no external force conditions, and the resection path and the incision location are uploaded to the cloud server.

2. The method according to claim 1, wherein the three-dimensional model reconstruction based on the prone MRI images, including at least extracting the skin model and tumor model of the affected breast, comprises: Based on preset MRI values ​​and target boundaries, a point set of the affected breast region is extracted from the prone MRI image. Based on the point set of the affected breast region, different MRI values ​​were set to filter and extract the tumor point set, blood vessel point set, and mammary gland point set, and three-dimensional reconstruction was performed using triangulation method to extract at least the skin model and the tumor model of the affected breast.

3. The method according to claim 1, wherein calculating the corresponding affected breast skin model and tumor model under no external force state based on the affected breast skin model and the tumor model includes: Material parameters were set for the affected breast skin model and the tumor model; The gravitational potential energy is superimposed onto the strain energy function of the material parameters of each model to construct a total strain energy function that includes elastic strain energy and gravitational potential energy. The displacement field of the affected breast skin and tumor tissue from a state with gravity to a state without external force is calculated based on the total strain energy function. The initial geometry of each model is corrected using the displacement field to obtain the corresponding skin model and tumor model of the affected breast under no external force conditions.

4. The method according to claim 3, wherein calculating the displacement field of the affected breast skin and tumor tissue from a state under gravity to a state without external force based on the total strain energy function includes: The momentum balance equation, which includes the contribution of gravity, is derived from the total strain energy function, and the momentum balance equation is transformed into an equivalent weak integral form. The affected breast skin and tumor tissue domain are divided into a finite number of units in the equivalent weak integral form, and a displacement shape function is constructed on each unit. Based on the equivalent weak integral form and the displacement shape function, a set of nonlinear algebraic equations is obtained in which the global nodal internal forces and global nodal external forces are equal among the elements in the affected breast skin and tumor tissue domain. The nonlinear algebraic equations are solved iteratively to obtain the displacement field of the affected breast skin and tumor tissue from a state with gravity to a state without external force.

5. The method according to claim 1, wherein determining the tumor resection path based on the direction vector of the shortest distance point between the tumor model and the affected breast skin model under no external force conditions comprises: The resection direction is obtained based on the direction vector of the shortest distance point between the tumor model and the skin model of the affected breast under no external force conditions; By tracing back to the resection direction and then to the surface of the affected breast skin model in the absence of external force, a continuous set of surface entry points is obtained to determine the resection path.

6. The method according to claim 5, wherein obtaining the resection direction based on the direction vector of the shortest distance point between the tumor model and the affected breast skin model under no external force conditions includes: Starting from the geometric center point of the tumor model under no external force condition, calculate the shortest distance point to the surface of the skin model of the affected breast under no external force condition; The resection direction is obtained based on the direction vector from the geometric center point to the shortest distance point.

7. The method of claim 6, wherein tracing backwards from the excision direction to the surface of the affected breast skin model in the absence of external force to obtain a continuous set of surface entry points to determine the excision path comprises: Calculate the vertical projection of the tumor model in the absence of external force onto the resection direction; Based on the boundary points of the vertical projection, trace them in the reverse direction of the excision to the surface of the affected breast skin model under the state of no external force, and obtain a continuous set of surface entry points. The cutting path is determined based on the polygonal channel formed by the set of surface entry points.

8. The method of claim 7, further comprising: In response to the intersection of the resection path and the vascular model or the minimum distance between the target point on the resection path and the vascular model being less than a distance threshold, the first intersection volume of the resection path and the vascular model after expanding the target region outward is calculated; The surface point set of the cutting path is rotated by a differential angle in multiple orthogonal planes to obtain the corresponding rotated cutting path. Calculate the second intersection volume between each rotated resection path and the target region after the blood vessel model is expanded outward; The rotational cut-off path corresponding to the largest difference between the second intersection volume and the first intersection volume is selected. The resection path corresponding to the largest difference is rotated by a differential angle, and the new intersection volume and the new rotated resection path are calculated. The process continues until the intersection volume between the new rotated resection path and the target region after the blood vessel model is expanded outward is zero, thus obtaining the final resection path.

9. An electronic device, comprising: processor; A memory storing computer instructions for determining the tumor resection path in breast-conserving surgery planning for breast cancer, wherein when the computer instructions are executed by a processor, the method according to any one of claims 1-8 is implemented.

10. A computer-readable storage medium having stored thereon computer program instructions for determining a tumor resection path in breast-conserving surgery planning for breast cancer, the computer program instructions, when executed by one or more processors, causing the method according to any one of claims 1-8 to be implemented.

Images