Pelvic organ model construction method and system based on MRI (Magnetic Resonance Imaging) three-dimensional reconstruction technology, and medium
By constructing a pelvic organ model based on MRI three-dimensional reconstruction technology, the limitations of multi-dimensional stereotactic analysis of pelvic floor diseases in the existing technology are solved, and precise biomechanical simulation and prolapse displacement analysis of pelvic organs are realized.
Patent Information
- Application Number
- CN202510440467.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-09
- Publication Date
- 2025-07-25
AI Technical Summary
In the prior art, the plane imaging mode is single, and multi-dimensional stereo dynamic analysis of female pelvic floor diseases cannot be achieved, resulting in significant limitations in the evaluation of pelvic organ prolapse type, mechanical coupling relationship between organs, and dynamic anatomical variation.
Through the MRI three-dimensional reconstruction technology, a pelvic organ model is constructed, including obtaining MRI data, tissue segmentation, point cloud data generation, solid model construction and cavity model assembly, and bladder, uterine, vagina and rectal models that meet the human anatomical structure, realizing multi-dimensional stereo dynamic analysis.
It provides an accurate pelvic organ model, which can seamlessly connect to finite element analysis software, perform biomechanical behavior simulation and pelvic organ prolapse displacement simulation, improving the accuracy of diagnosis and treatment.
Smart Images

Figure CN120374846A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of medical image processing and three-dimensional modeling, and specifically relates to a method, system and medium for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology. Background Art
[0002] Female pelvic organ prolapse (FPOP) refers to the downward displacement of one or more of the anterior vaginal wall, posterior vaginal wall, uterus or vaginal apex, and is a common pelvic floor dysfunction disease that affects the quality of life of middle-aged and elderly women. At present, surgical operation is the main treatment method for FPOP, but the problem of its recurrence rate has long troubled clinical practice. Long-term illness and repeated treatment have caused heavy economic burden and psychological pressure to patients.
[0003] Accurate diagnosis and early intervention are the keys to reducing the negative impacts of female pelvic floor diseases on health and economy. Although B-ultrasound and two-dimensional MRI (magnetic resonance imaging) are currently used for imaging judgment of the degree of prolapse, limited by the single-plane imaging mode, multi-dimensional stereoscopic dynamic analysis cannot be achieved, resulting in significant limitations in the evaluation of prolapse types, mechanical coupling relationships between organs and dynamic anatomical variations.
[0004] Therefore, there is an urgent need to provide a method, system and medium for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology to achieve multi-dimensional stereoscopic dynamic analysis of pelvic organ prolapse. Summary of the Invention
[0005] The purpose of the present invention is to provide a method, system and medium for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology, effectively solving the technical problem that the single-plane imaging mode in the prior art cannot achieve multi-dimensional stereoscopic dynamic analysis.
[0006] To solve the above technical problems, in the first aspect, the present invention provides a method for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology, including the following steps:
[0007] Obtain MRI data for constructing a pelvic organ model;
[0008] Perform pelvic organ tissue segmentation based on the MRI data to obtain three-dimensional point cloud data of pelvic organ tissues; wherein, the three-dimensional point cloud data of pelvic organ tissues includes three-dimensional point cloud data of the bladder, three-dimensional point cloud data of the uterus, three-dimensional point cloud data of the endometrium, three-dimensional point cloud data of the vagina, and three-dimensional point cloud data of the rectum;
[0009] Perform three-dimensional modeling based on the three-dimensional point cloud data of pelvic organ tissues to obtain a solid model of pelvic organ tissues; among them, the solid model of pelvic organ tissues includes a bladder solid model, a uterine solid model, an endometrial solid model, a vaginal solid model, and a rectal solid model;
[0010] Construct a bladder cavity model, a muscular uterine cavity model, a vaginal cavity model, and a rectal cavity model based on the solid model of pelvic organ tissues;
[0011] Perform origin assembly on the bladder cavity model, the muscular uterine cavity model, the vaginal cavity model, and the rectal cavity model to obtain a pelvic organ model.
[0012] Preferably, perform segmentation of pelvic organ tissues based on MRI data to obtain three-dimensional point cloud data of pelvic organ tissues, which specifically includes the following sub-steps:
[0013] Based on the distribution characteristics of the MRI image gray histogram, respectively outline the initial contours of the bladder, uterus, endometrium, vagina, and rectum, and strengthen the density difference boundaries between the bladder wall, uterine muscle layer, endometrium, vaginal wall, and rectal wall and the surrounding tissues to obtain a bladder mask, a uterine mask, an endometrial mask, a vaginal mask, and a rectal mask that contain the complete anatomical features of the organs;
[0014] Perform threshold segmentation, region growing, interlayer contour correction, and mask smoothing processing based on the bladder mask, uterine mask, endometrial mask, vaginal mask, and rectal mask to obtain bladder three-dimensional point cloud data, uterine three-dimensional point cloud data, endometrial three-dimensional point cloud data, vaginal three-dimensional point cloud data, and rectal three-dimensional point cloud data.
[0015] Preferably, in the step of respectively outlining the initial contours of the bladder, uterus, endometrium, vagina, and rectum based on the distribution characteristics of the MRI image gray histogram and strengthening the density difference boundaries between the bladder wall, uterine muscle layer, endometrium, vaginal wall, and rectal wall and the surrounding tissues, the following steps are further included:
[0016] Extract the wall thickness quantitative parameters of the bladder wall, vaginal wall, and rectal wall respectively, and calculate the average wall thickness of the bladder wall, vaginal wall, and rectal wall respectively to obtain the average bladder wall thickness, average vaginal wall thickness, and average rectal wall thickness.
[0017] Preferably, perform three-dimensional modeling based on the three-dimensional point cloud data of pelvic organ tissues to obtain a solid model of pelvic organ tissues, which specifically includes the following sub-steps:
[0018] Perform operations such as nail elimination, point cloud relaxation, noise filtering, hole repair, and surface smoothing on the three-dimensional point cloud data of pelvic organ tissues to form a triangular mesh model of pelvic organ tissues; among them, the triangular mesh model of pelvic organ tissues includes a bladder triangular mesh model, a uterine triangular mesh model, a uterine triangular mesh model, a vaginal triangular mesh model, and a rectal triangular mesh model;
[0019] Perform non-uniform rational B-spline surface reconstruction based on the triangular mesh model of pelvic organ tissues to obtain a solid model of pelvic organ tissues.
[0020] Preferably, in the step of constructing a bladder cavity model, a muscular uterine cavity model, a vaginal cavity model, and a rectal cavity model based on the solid model of pelvic organ tissues, the steps of constructing the bladder cavity model are as follows:
[0021] Perform plane cutting on the bladder solid model to remove the distal end of the urethral orifice in the bladder solid model to obtain a bladder solid cutting model;
[0022] Perform shelling on the bladder solid cutting model based on the average value of the bladder wall thickness to obtain an initial bladder cavity model with an open distal end of the urethral orifice and a semi-closed bladder main body;
[0023] Perform operations such as nail elimination, mesh relaxation, noise filtering, surface smoothing, non-uniform rational B-spline surface reconstruction, grid construction, and geometric verification on the distal end of the urethral orifice in the initial bladder cavity model to obtain a bladder cavity model.
[0024] Preferably, in the step of constructing a bladder cavity model, a muscular uterine cavity model, a vaginal cavity model, and a rectal cavity model based on the solid model of pelvic organ tissues, the steps of constructing the muscular uterine cavity model are as follows:
[0025] Perform origin assembly on the uterine solid model and the endometrial solid model to obtain an initial muscular uterine solid model;
[0026] Construct a first model copy of the initial muscular uterine solid model based on the initial muscular uterine solid model to obtain the first model copy of the initial muscular uterine solid model;
[0027] Perform a Boolean subtraction operation on the initial muscular uterine solid model and the first model copy of the initial muscular uterine solid model to obtain a muscular uterine cavity model with the anatomical morphology of the uterine myometrium.
[0028] Preferably, in the step of constructing a bladder cavity model, a muscular uterine cavity model, a vaginal cavity model, and a rectal cavity model based on the solid model of pelvic organ tissues, the steps of constructing the vaginal cavity model are as follows:
[0029] Perform planar cutting on the vaginal solid model, and remove the lower end of the vaginal solid model to obtain a vaginal solid cutting model;
[0030] Perform shelling on the vaginal solid cutting model based on the average vaginal wall thickness to obtain a first initial model of the vaginal cavity body that is fully enclosed at the upper end and open at the lower end;
[0031] Perform nail elimination, mesh relaxation, noise filtering, surface smoothing, non-uniform rational B-spline surface reconstruction, grid construction, and geometric verification on the lower end of the first initial model of the vaginal cavity body in sequence to obtain a second initial model of the vaginal cavity body;
[0032] Take the initial muscular uterine solid model and the second initial model of the vaginal cavity body for origin assembly to obtain a third initial model of the vaginal cavity body;
[0033] Construct a second model copy of the initial muscular uterine solid model based on the initial muscular uterine solid model to obtain a second model copy of the initial muscular uterine solid;
[0034] Perform a Boolean subtraction operation based on the third initial model of the vaginal cavity body and the second model copy of the initial muscular uterine solid to obtain a vaginal cavity body model with a cervical anatomical surrounding structure at the upper end and open at the lower end.
[0035] Preferably, in the step of constructing the bladder cavity body model, the muscular uterine cavity body model, the vaginal cavity body model, and the rectal cavity body model based on the pelvic organ tissue solid model, the steps of constructing the rectal cavity body model are as follows:
[0036] Perform planar cutting on the rectal solid model, and remove the starting end and the end of the rectal solid model to obtain a rectal solid cutting model;
[0037] Perform shelling on the rectal solid cutting model based on the average rectal wall thickness to obtain an initial rectal cavity body model that is open at the starting end and the end;
[0038] Perform nail elimination, mesh relaxation, noise filtering, surface smoothing, non-uniform rational B-spline surface reconstruction, grid construction, and geometric verification on the starting end and the end of the initial rectal cavity body model in sequence to obtain a rectal cavity body model.
[0039] In a second aspect, the present invention also provides a pelvic organ model construction system based on MRI three-dimensional reconstruction technology, including an MRI data acquisition module, a pelvic organ tissue segmentation module, a pelvic organ tissue three-dimensional modeling module, a pelvic organ tissue cavity body construction module, and a pelvic organ assembly module; wherein,
[0040] The MRI data acquisition module is used to acquire MRI data for constructing the pelvic organ model;
[0041] A pelvic organ tissue segmentation module, which is used to segment pelvic organ tissues based on MRI data to obtain three-dimensional point cloud data of pelvic organ tissues; among them, the three-dimensional point cloud data of pelvic organ tissues includes three-dimensional point cloud data of the bladder, three-dimensional point cloud data of the uterus, three-dimensional point cloud data of the endometrium, three-dimensional point cloud data of the vagina, and three-dimensional point cloud data of the rectum;
[0042] A pelvic organ tissue three-dimensional modeling module, which is used to perform three-dimensional modeling based on the three-dimensional point cloud data of pelvic organ tissues to obtain a solid model of pelvic organ tissues; among them, the solid model of pelvic organ tissues includes a solid model of the bladder, a solid model of the uterus, a solid model of the endometrium, a solid model of the vagina, and a solid model of the rectum;
[0043] A pelvic organ tissue cavity construction module, which is used to construct a bladder cavity model, a muscular uterine cavity model, a vaginal cavity model, and a rectal cavity model based on the solid model of pelvic organ tissues;
[0044] A pelvic organ assembly module, which is used to perform origin assembly on the bladder cavity model, the muscular uterine cavity model, the vaginal cavity model, and the rectal cavity model to obtain a pelvic organ model.
[0045] In a third aspect, the present invention also provides a computer storage medium. The computer storage medium stores a computer program, and the computer program includes program instructions. When the program instructions are executed by a processor, the processor is caused to execute the method for constructing a pelvic organ model based on the MRI three-dimensional reconstruction technology described in the first aspect.
[0046] Adopting the technical solution provided by the present invention, compared with the prior art, it has the following advantages:
[0047] A method for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology provided by the present invention. First, the MRI data of a patient is obtained as the data basis for constructing the pelvic organ model. Based on the MRI data, pelvic organ tissue segmentation is performed to generate three-dimensional point cloud data of the bladder, uterus, endometrium, vagina, and rectum with anatomical authenticity. Secondly, based on the three-dimensional point cloud data of the bladder, uterus, endometrium, vagina, and rectum, a regularized polygon mesh is constructed through point cloud denoising, hole repair, and smoothing processing. Through processes such as surface exploration, contour line editing, and grid construction, the irregular free-form surface is transformed into a uniform NURBS surface patch to improve the surface smoothness and geometric accuracy of the model, thereby generating a bladder solid model, uterus solid model, endometrium solid model, vagina solid model, and rectum solid model containing complete geometric topology information. Then, by performing nested reduction on the bladder solid model, uterus solid model, endometrium solid model, vagina solid model, and rectum solid model, a bladder cavity model, muscular uterine cavity model, vaginal cavity model, and rectum cavity model with complex cavity and pipeline structures are formed, so that the pelvic organ model obtained by subsequent assembly is closer to the real anatomical structure. Finally, by performing origin assembly on the bladder cavity model, muscular uterine cavity model, vaginal cavity model, and rectum cavity model, a pelvic organ model that conforms to the real anatomical structure of the human body and the precise organ adjacency relationship is obtained. In subsequent applications, this pelvic organ model can not only be seamlessly docked with finite element analysis software to simulate biomechanical behaviors such as stress distribution of pelvic floor muscle groups and contact mechanics between organs, but also can carry out dynamic evolution analysis of anatomical morphology such as pelvic organ prolapse displacement simulation through a three-dimensional visualization platform. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] In order to more clearly illustrate the technical solutions of the prior art and the embodiments of the present application, the drawings required for describing the prior art and the embodiments of the present application will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments recorded in the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.
[0049] Figure 1 It is a schematic diagram of the overall process of a method for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology in Embodiment 1.
[0050] Figure 2 It is a schematic diagram of the process for constructing a bladder cavity model in Embodiment 1.
[0051] Figure 3Schematic diagram of the process for constructing a muscular uterine cavity model in Example 1.
[0052] Figure 4 Schematic diagram of the process for constructing a vaginal cavity model in Example 1.
[0053] Figure 5 Schematic diagram of the process for constructing a rectal cavity model in Example 1. Detailed implementation manners
[0054] To better understand the purpose, structure and function of the present invention, the following further describes in detail a method, system and medium for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology provided by the present invention with reference to the accompanying drawings, so that those skilled in the art can better understand the present invention and be able to implement it, but the examples given are not intended to limit the present invention.
[0055] Example 1
[0056] Please refer to Figure 1 As shown, a method for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology is provided in an embodiment of the present application, including the following steps:
[0057] Step S1: Obtain MRI data for constructing a pelvic organ model.
[0058] Specifically, the purpose of this step is to extract the Digital Imaging and Communications in Medicine (DICOM) data generated by a patient's pelvic MRI scan, and import the MRI DICOM data into a medical image processing system for image scanning, providing a data basis for constructing a pelvic organ model. The medical image processing system used in the embodiment of the present application is Mimics (Materialise's interactive medical image control system). After the Mimics system automatically completes image registration and preprocessing, it synchronously generates three mutually related coronal, transverse and sagittal view windows on the interactive interface. Among them, the transverse view window displays the original MRI scan data.
[0059] Step S2: Perform pelvic organ tissue segmentation based on the MRI data to obtain three-dimensional point cloud data of pelvic organ tissues; among them, the three-dimensional point cloud data of pelvic organ tissues includes three-dimensional point cloud data of the bladder, three-dimensional point cloud data of the uterus, three-dimensional point cloud data of the endometrium, three-dimensional point cloud data of the vagina, and three-dimensional point cloud data of the rectum.
[0060] Specifically, the purpose of this step is to separate the tissue structures of pelvic organs (such as the bladder, uterus, endometrium, vagina, and rectum) from the surrounding tissues and background in the overall data of MRI images, and obtain the three-dimensional point cloud data of the bladder, uterus, endometrium, vagina, and rectum, providing a basis for subsequent precise modeling. Further, using the cross-sectional, sagittal, and coronal plane windows interconnected by the Mimics system, the anatomical boundaries of the bladder, vagina, and rectum are framed layer by layer, and through layer-by-layer observation, the myometrium and endometrial structures of the uterus are ensured to be fully developed. Specifically, it includes the following sub-steps:
[0061] Step S21: Based on the distribution characteristics of the MRI image gray histogram, the initial contours of the bladder, uterus, endometrium, vagina, and rectum are outlined respectively, and the density difference boundaries between the bladder wall, myometrium, endometrium, vaginal wall, and rectal wall and the surrounding tissues are strengthened to obtain the bladder mask, uterus mask, endometrium mask, vagina mask, and rectal mask containing the complete anatomical features of the organs.
[0062] Specifically, by analyzing the peak and valley positions of the pelvic MRI image gray histogram, the gray boundaries between different tissues can be determined to outline the initial contours of the bladder, uterus, endometrium, vagina, and rectum. By strengthening the density difference boundaries (i.e., the imaging gray thresholds) between the bladder wall, myometrium, endometrium, vaginal wall, and rectal wall and the surrounding tissues, the tissue structures of pelvic organs can be more accurately segmented from the surrounding tissues, which is beneficial to more precisely constructing the models of the internal structures and external contours of each tissue structure in the subsequent process to be closer to the real anatomical structure. During this process, the wall thickness quantitative parameters of the bladder wall, vaginal wall, and rectal wall are extracted respectively, and the average wall thicknesses of the bladder wall, vaginal wall, and rectal wall are calculated respectively to obtain the average bladder wall thickness, average vaginal wall thickness, and average rectal wall thickness. In addition, during the construction of the bladder mask, uterus mask, endometrium mask, vagina mask, and rectal mask, the multi-plane views are synchronously verified to ensure the continuity of the organ contours in three-dimensional space. It should be noted that when outlining the endometrial boundary, special attention should be paid to the anatomical termination point, that is, the endometrium mask should terminate at the level of the internal os of the cervix. At the same time, according to the morphological characteristics of the endometrium at the end of the internal os of the cervix, 1-2 consecutive tomograms are drawn downward to maintain the anatomical continuity of the endometrial model in three-dimensional space, and the spatial nesting relationship between the endometrium mask and the uterus mask is verified synchronously by the three views to ensure the structural integrity of the subsequent nested muscular uterine cavity model.
[0063] Furthermore, the calculation process of the average bladder wall thickness, average vaginal wall thickness, and average rectal wall thickness strictly follows the double-operator independent same-layer cross-validation measurement specification. The measurement results of the two operators are subjected to consistency verification through the Intraclass Correlation Coefficient (ICC) to eliminate the differences between operators. For example, operators A and B respectively perform three repeated measurements on the bladder wall, vaginal wall, and rectal wall at the same imaging level, and obtain the measurement value sets of the same target, that is, the measurement value set of operator A {x A1 ,x A2 ,x A3} and the measurement value set of operator B {x B1 ,x B2 ,x B3}, and the intraclass correlation coefficient is calculated through a two-way mixed-effect model. The calculation formula is as follows:
[0064]
[0065] If the ICC verification is passed, that is, ICC≥0.75, the differences between operators can be ignored, and the measurement means of operators A and B in the three measurements are calculated respectively. The calculation formulas are as follows:
[0066]
[0067] Secondly, the wall thickness mean value (unit: mm, numerical accuracy reserved to two decimal places) is obtained based on the three measurement means. The calculation formula is as follows:
[0068]
[0069] By establishing a non-linear regression model between the MRI image gray value and the tissue thickness, it is used to realize the dynamic mapping of the shelling parameters, rather than using fixed empirical values or regional means in the traditional method, effectively solving the technical problem of the cavity wall thickness distortion caused by uneven thickness distribution in the prior art.
[0070] Step S22: Based on the bladder mask, uterine mask, endometrial mask, vaginal mask, and rectal mask, perform threshold segmentation, region growing, interlayer contour correction, and mask smoothing processing to obtain bladder three-dimensional point cloud data, uterine three-dimensional point cloud data, endometrial three-dimensional point cloud data, vaginal three-dimensional point cloud data, and rectal three-dimensional point cloud data.
[0071] Specifically, the Otsu algorithm is used to achieve threshold segmentation of pelvic tissues; the eight-neighborhood region growing algorithm is used to optimize the boundaries of pelvic organ tissues obtained by threshold segmentation, making the boundaries of pelvic organ tissues more accurate and continuous. The B-spline curve is used to correct the interlayer contours of pelvic tissues, making the contours more in line with the actual anatomical structure. By smoothing the mask, the three-dimensional point cloud data of the bladder, uterus, endometrium, vagina, and rectum after segmentation can form a smoother surface during subsequent modeling.
[0072] Step S3: Based on the three-dimensional point cloud data of pelvic organ tissues, three-dimensional modeling is performed to obtain a solid model of pelvic organ tissues; among them, the solid model of pelvic organ tissues includes a solid model of the bladder, a solid model of the uterus, a solid model of the endometrium, a solid model of the vagina, and a solid model of the rectum. Specifically, it includes the following sub-steps:
[0073] The three-dimensional point cloud data of pelvic organ tissues is subjected to spike elimination, point cloud relaxation, noise filtering, hole repair, and surface smoothing to form a triangular mesh model of pelvic organ tissues; among them, the triangular mesh model of pelvic organ tissues includes a triangular mesh model of the bladder, a triangular mesh model of the uterus, a triangular mesh model of the uterus, a triangular mesh model of the vagina, and a triangular mesh model of the rectum.
[0074] Specifically, during the process of three-dimensional point cloud data acquisition, some isolated and protruding "spike" points may be generated due to equipment errors, abnormal reflections, or other interference factors, which will affect the accuracy of subsequent model construction. Spike elimination is used to detect and remove these abnormal data points to prevent these abnormal points from causing local distortion of the model during subsequent processing such as surface reconstruction, making the shape of the solid model of pelvic organ tissues more in line with the actual anatomical structure. Point cloud relaxation is used to make the distribution of the three-dimensional point cloud data of pelvic organ tissues more coherent in space, which can better reflect the continuous surface characteristics of pelvic organ tissues. By performing noise filtering, the three-dimensional point cloud data of pelvic organ tissues can show the details of pelvic organ tissues more clearly, such as the thickness of the organ wall and the boundaries between tissues, ensuring that the constructed pelvic organ model can accurately reflect the true shape and structure of the tissues. By repairing the holes in the three-dimensional point cloud data of pelvic organ tissues, it is used to improve the topological structure of the three-dimensional point cloud data of pelvic organ tissues, ensuring that a complete pelvic organ model can be constructed and avoiding discontinuities on the model surface. By performing surface smoothing, it is used to adjust the shape of the surface, eliminate non-smooth factors such as wrinkles and edges on the surface, and make the surface more natural and smooth. The three-dimensional point cloud data of pelvic organ tissues after the above-mentioned processing is converted into a high-precision triangular mesh to form a triangular mesh model of pelvic organ tissues, and geometric defects are eliminated through contour line editing and surface patch reconstruction.
[0075] Perform non-uniform rational B-spline surface reconstruction based on the pelvic organ tissue triangular mesh model to obtain the pelvic organ tissue solid model.
[0076] Specifically, by using the Non-Uniform Rational B-Spline (NURBS) algorithm to generate a parametric surface, not only can the shape of the surface be flexibly adjusted to match the actual contour of the pelvic organ tissue, so as to generate personalized surfaces according to the anatomical structure differences of the pelvic organs of different patients, construct pelvic organ models that conform to individual characteristics, and improve the construction accuracy of the pelvic organ models, but also it can facilitate the subsequent interaction between the pelvic organ model and finite element analysis software for mechanical analysis. Finally, the fitting surface is generated through grille construction and topology optimization processing to obtain the pelvic organ tissue solid model.
[0077] Step S4: Construct a bladder cavity model, a muscular uterine cavity model, a vaginal cavity model, and a rectal cavity model based on the pelvic organ tissue solid model. Among them,
[0078] Please refer to Figure 2 As shown, the steps to construct the bladder cavity model are as follows:
[0079] Step S41-1: Perform plane cutting on the bladder solid model, and remove the distal end of the urethral orifice in the bladder solid model to obtain the bladder solid cutting model.
[0080] Specifically, first import the bladder solid model into Geomagic Wrap (3D reverse engineering software), and by adjusting the spatial pose parameters (translation and rotation transformation) of the global coordinate system, make the cutting plane parallel to the anatomical position of the distal end of the urethral orifice in the bladder solid model. Subsequently, use the plane section function to perform geometric cutting on the distal end of the urethral orifice along the preset plane, and remove the irregular redundant area to obtain the bladder solid cutting model, whose topological structure meets the anatomical morphological characteristics of the bladder cavity.
[0081] Step S41-2: Perform shelling on the bladder solid cutting model based on the average bladder wall thickness to obtain an initial bladder cavity model with an open distal end of the urethral orifice and a semi-closed bladder body.
[0082] Specifically, set the shelling parameters based on the average bladder wall thickness measured by the patient's MRI, and perform shelling on the bladder solid cutting model to obtain an initial bladder cavity model with an open distal end of the urethral orifice and a semi-closed bladder body.
[0083] Step S41-3: Perform nail elimination, mesh relaxation, noise filtering, surface smoothing, non-uniform rational B-spline surface reconstruction, grille construction, and geometric verification on the distal end of the urethral orifice in the initial bladder cavity model to obtain the bladder cavity model.
[0084] Specifically, for the geometric sharp edge formed by cutting the end plane of the urethral orifice in the initial model of the bladder cavity, nail elimination, mesh relaxation, noise filtering, and surface smoothing are sequentially performed to make the morphological transition conform to the urethral anatomical characteristics. Then, through the non-uniform rational B-spline surface reconstruction technology, topological optimization and surface patch reconstruction and repair are carried out on the contour line. Finally, the cavity solid modeling is completed through grille structure generation and geometric verification, and the initial model of the bladder cavity with the end of the urethral orifice open and the main body of the bladder semi-closed is obtained.
[0085] Please refer to Figure 3 as shown in the figure, the steps for constructing the muscular uterine cavity model are as follows:
[0086] Step S42-1: Assemble the origin of the uterine solid model and the endometrial solid model to obtain the initial solid model of the muscular uterus.
[0087] Specifically, use SolidWorks 3D software to perform origin assembly on the uterine solid model and the endometrial solid model, and apply coincidence constraints at the same time to achieve the precise nesting of anatomical structures in three-dimensional space, and obtain the initial solid model of the muscular uterus.
[0088] Step S42-2: Based on the initial solid model of the muscular uterus, construct the first model copy of the initial solid model of the muscular uterus to obtain the first model copy of the initial solid model of the muscular uterus.
[0089] Specifically, use SolidWorks 3D software to construct the first model copy of the initial solid model of the muscular uterus to obtain the first model copy of the initial solid model of the muscular uterus, which is used as the endometrial solid model copy for Boolean subtraction operation.
[0090] Step S42-3: Perform a Boolean subtraction operation based on the initial solid model of the muscular uterus and the first model copy of the initial solid model of the muscular uterus to obtain the muscular uterine cavity model with the anatomical morphology of the uterine myometrium.
[0091] Specifically, select the initial solid model of the muscular uterus as the main entity, select the endometrial solid model in the first model copy of the initial solid model of the muscular uterus as the entity to be deleted, perform a Boolean logic operation, and obtain the muscular uterine cavity model with the anatomical morphology of the uterine myometrium. Its irregular wall thickness characteristics conform to clinical anatomical characteristics, and the geometric accuracy meets the requirements of subsequent biomechanical simulation analysis.
[0092] Please refer to Figure 4 as shown in the figure, the steps for constructing the vaginal cavity model are as follows:
[0093] Step S43-1: Based on the vaginal solid model, perform plane cutting processing to remove the lower end of the vaginal solid model to obtain the vaginal solid cutting model.
[0094] Specifically, first, import the vaginal solid model into Geomagic Wrap. By adjusting the spatial pose of the global coordinate system (including translation and rotation transformation), make the cutting plane spatially parallel to the anatomical position at the lower end of the vaginal solid model. Subsequently, apply the plane section function to geometrically cut the lower end of the vagina along the preset plane, remove the irregular redundant areas, and obtain the vaginal solid cutting model.
[0095] Step S43-2: Perform a shelling process on the vaginal solid cutting model based on the average vaginal wall thickness to obtain the first initial model of the vaginal cavity body with a fully enclosed upper segment and an open lower segment.
[0096] Specifically, set the shelling parameters based on the average vaginal wall thickness measured by the patient's MRI, and perform a shelling process on the vaginal solid cutting model to obtain the first initial model of the vaginal cavity body with a fully enclosed upper segment and an open lower segment.
[0097] Step S43-3: Successively perform nail elimination, mesh relaxation, noise filtering, surface smoothing, non-uniform rational B-spline (NURBS) surface reconstruction, grid construction, and geometric verification on the lower end of the first initial model of the vaginal cavity body to obtain the second initial model of the vaginal cavity body.
[0098] Specifically, for the geometric sharp edges formed by plane cutting at the open end of the lower vagina after plane cutting, successively perform nail elimination, mesh relaxation, noise removal, and surface smoothing to make its transition form conform to the human anatomical characteristics. Then, adopt the non-uniform rational B-spline surface reconstruction technology to achieve high-precision surface modeling through contour line topology optimization, surface patch reconstruction, and repair. Finally, through grid structure generation and geometric verification, obtain the second initial model of the vaginal cavity body.
[0099] Step S43-4: Take the initial muscular uterine solid model and the second initial model of the vaginal cavity body for origin assembly to obtain the third initial model of the vaginal cavity body.
[0100] Specifically, use SolidWorks 3D software to perform origin assembly on the initial muscular uterine solid model and the second initial model of the vaginal cavity body to obtain the third initial model of the vaginal cavity body. It should be noted that this third initial model of the vaginal cavity body is a composite assembly with geometric penetration and interference characteristics. Due to the unoptimized anatomical structure, there is an abnormal spatial penetration phenomenon of the uterine-endometrial nested body and the vaginal cavity, that is, the penetration / interference phenomenon, and it is temporarily unable to accurately represent the cervical-vaginal anatomical wrapping relationship, providing basic model data for subsequent anatomical morphology correction processing.
[0101] Step S43-5: Based on the initial muscular uterine solid model, construct a second model copy of the initial muscular uterine solid model to obtain the second model copy of the initial muscular uterine solid model.
[0102] Specifically, a second model copy of the initial solid model of the muscular uterus is constructed using SolidWorks 3D software to obtain the second model copy of the initial solid model of the muscular uterus.
[0103] Step S43-6: Perform a Boolean subtraction operation based on the third initial model of the vaginal cavity body and the second model copy of the initial solid model of the muscular uterus to obtain a vaginal cavity body model with a cervical anatomical wrapping structure in the upper segment and an open lower segment.
[0104] Specifically, select the third initial model of the vaginal cavity body as the main entity, select the second model copy of the initial solid model of the muscular uterus as the entity to be subtracted, perform a Boolean logic operation to eliminate the penetration area of the uterus-endometrium structure in the vaginal cavity, and finally obtain an optimized model that conforms to the cervical anatomical wrapping relationship, that is, obtain a vaginal cavity body model with a cervical anatomical wrapping structure in the upper segment and an open lower segment. This vaginal cavity body model basically conforms to the true human vaginal anatomical structure and can provide a basis for biomechanical analysis or surgical planning to provide an accurate anatomical model. This step adopts the priority deletion rule of the anatomical wrapping relationship, that is, the cervical wrapping area is preferentially retained, replacing the non-discriminatory deletion mode in the traditional Boolean operation, and avoiding the loss of anatomical features (such as the destruction of the cervical fornix morphology) caused by the single logic of geometric interference elimination in the existing method, and increasing the retention rate of the key anatomical features of the cavity model to 98%.
[0105] Please refer to Figure 5 as shown in the figure, the steps to construct the rectal cavity body model are as follows:
[0106] Step S44-1: Perform a plane cutting process on the rectal solid model to remove the starting end and the ending end of the rectal solid model to obtain a rectal solid cutting model.
[0107] Specifically, first import the rectal solid model into Geomagic Wrap software, and by adjusting the spatial pose parameters (translation and rotation transformation) of the global coordinate system, make the cutting plane parallel to the anatomical starting segment of the rectal solid model in space. Subsequently, apply the plane section function to perform geometric cutting on the starting and ending ends of the rectum along the preset plane to remove irregular redundant areas and obtain a rectal solid cutting model. Its geometric features meet the anatomical morphological requirements of the rectal cavity.
[0108] Step S44-2: Perform a shelling process on the rectal solid cutting model based on the average rectal wall thickness to obtain an initial model of the rectal cavity body with open starting and ending ends.
[0109] Specifically, set the shelling parameters based on the average rectal wall thickness measured by the patient's MRI and perform a shelling process on the rectal solid cutting model to obtain an initial model of the rectal cavity body with open starting and ending ends.
[0110] Step S44-3: Perform nail-like object elimination, mesh relaxation, noise filtering, surface smoothing, non-uniform rational B-spline (NURBS) surface reconstruction, grille construction, and geometric verification processing on the starting end and the ending end of the initial rectal cavity model in sequence to obtain the rectal cavity model.
[0111] Specifically, for the geometric sharp edges formed after planar cutting of the starting end and the ending end of the initial rectal cavity model, perform nail-like object elimination, mesh relaxation, noise filtering, and surface smoothing processing in sequence to make the morphological transition conform to the rectal anatomical characteristics. Then, adopt the NURBS surface reconstruction technology to achieve high-precision surface modeling through contour line topology optimization, surface patch reconstruction and repair. Finally, obtain the rectal cavity model through grille structure generation and geometric verification.
[0112] Step S5: Perform origin assembly on the bladder cavity model, the muscular uterine cavity model, the vaginal cavity model, and the rectal cavity model to obtain the pelvic organ model.
[0113] Specifically, first import the bladder cavity model, the muscular uterine cavity model, the vaginal cavity model, and the rectal cavity model into the SolidWorks 3D software respectively. Then, call the parametric dynamic assembly constraint library, check and correct the positional relationships of each organ according to the anatomical variation parameters (such as the bladder-vagina inclination angle, the depth of the rectouterine pouch) defined by the clinical imaging data. When necessary, apply distance constraints and angle constraints based on anatomical landmark points. Finally, generate a pelvic organ model with anatomical continuity, and this pelvic organ model conforms to the true human anatomical structure and the precise organ adjacency relationship. By establishing a dynamic assembly constraint library based on anatomical variation metrics (such as the dynamic range of the uterus-vagina inclination angle is 15°-30°), it supports one-key parameter-driven assembly, thus replacing the traditional manual adjustment case by case, improving the assembly efficiency of complex anatomical systems by 3-5 times, and covering more than 90% of the clinically common anatomical variation types, realizing the dynamic adaptation of multi-organ adjacency relationships. In addition, this pelvic organ model retains the NURBS surface parameters, material properties, and spatial constraint relationships through the STEP AP242 standard. Therefore, in subsequent applications, the female pelvic organ model constructed by this method can not only be seamlessly docked with finite element analysis software to realize the simulation of biomechanical behaviors such as the stress distribution of the pelvic floor muscles and the contact mechanics between organs, but also can carry out dynamic evolution analysis of anatomical morphology such as pelvic organ prolapse displacement simulation through a 3D visualization platform.
[0114] Combining the above steps, a method for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology provided in this embodiment is as follows. First, the MRI data of the patient is obtained as the data basis for constructing the pelvic organ model. Based on the MRI data, the pelvic organ tissues are segmented to generate bladder three-dimensional point cloud data, uterine three-dimensional point cloud data, endometrial three-dimensional point cloud data, vaginal three-dimensional point cloud data, and rectal three-dimensional point cloud data with anatomical authenticity. Secondly, based on the bladder three-dimensional point cloud data, uterine three-dimensional point cloud data, endometrial three-dimensional point cloud data, vaginal three-dimensional point cloud data, and rectal three-dimensional point cloud data, a regularized polygon mesh is constructed through point cloud denoising, hole repair, and smoothing processing. Through processes such as surface exploration, contour line editing, and grid construction, the irregular free-form surface is transformed into a uniform NURBS surface patch to improve the surface smoothness and geometric accuracy of the model, thereby generating a bladder solid model, uterine solid model, endometrial solid model, vaginal solid model, and rectal solid model containing complete geometric topology information. Then, by performing nested cutting on the bladder solid model, uterine solid model, endometrial solid model, vaginal solid model, and rectal solid model, a bladder cavity model, muscular uterine cavity model, vaginal cavity model, and rectal cavity model with complex cavity and pipeline structures are formed, so that the subsequent assembled pelvic organ model is closer to the real anatomical structure. Finally, by performing origin assembly on the bladder cavity model, muscular uterine cavity model, vaginal cavity model, and rectal cavity model, a pelvic organ model that conforms to the real anatomical structure of the human body and the precise organ adjacency relationship is obtained. In subsequent applications, this pelvic organ model can not only be seamlessly docked with finite element analysis software to realize biomechanical behavior simulations such as stress distribution of pelvic floor muscle groups and contact mechanics between organs, but also carry out dynamic evolution analysis of anatomical morphology such as pelvic organ prolapse displacement simulation through a three-dimensional visualization platform.
[0115] Embodiment 2
[0116] This embodiment provides a system for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology, including an MRI data acquisition module, a pelvic organ tissue segmentation module, a pelvic organ tissue three-dimensional modeling module, a pelvic organ tissue cavity construction module, and a pelvic organ assembly module; among them,
[0117] The MRI data acquisition module is used to acquire MRI data for constructing the pelvic organ model;
[0118] The pelvic organ tissue segmentation module is used to segment the pelvic organ tissues based on the MRI data to obtain pelvic organ tissue three-dimensional point cloud data; among them, the pelvic organ tissue three-dimensional point cloud data includes bladder three-dimensional point cloud data, uterine three-dimensional point cloud data, endometrial three-dimensional point cloud data, vaginal three-dimensional point cloud data, and rectal three-dimensional point cloud data;
[0119] A three-dimensional modeling module for pelvic organ tissues, which is used to perform three-dimensional modeling based on the three-dimensional point cloud data of pelvic organ tissues to obtain a solid model of pelvic organ tissues; among them, the solid model of pelvic organ tissues includes a bladder solid model, a uterine solid model, an endometrial solid model, a vaginal solid model, and a rectal solid model;
[0120] A cavity construction module for pelvic organ tissues, which is used to construct a bladder cavity model, a muscular uterine cavity model, a vaginal cavity model, and a rectal cavity model based on the solid model of pelvic organ tissues;
[0121] A pelvic organ assembly module, which is used to perform origin assembly on the bladder cavity model, the muscular uterine cavity model, the vaginal cavity model, and the rectal cavity model to obtain a pelvic organ model.
[0122] Embodiment Three
[0123] This embodiment provides a computer storage medium. The computer storage medium stores a computer program, and the computer program includes program instructions. When the program instructions are executed by a processor, the processor executes the method for constructing a pelvic organ model based on the MRI three-dimensional reconstruction technology described in Embodiment One.
[0124] Embodiment Four
[0125] This embodiment provides a computer device, which includes a memory, a processor, and a computer program stored on the memory and running on the processor. When the processor executes the program, it implements the method for constructing a pelvic organ model based on the MRI three-dimensional reconstruction technology as described in Embodiment One above.
[0126] It should be noted here that the steps involved in the above Embodiments Two, Three, and Four correspond to those in Embodiment One. For the specific implementation methods, please refer to the relevant description part of Embodiment One, which will not be elaborated here.
[0127] It can be understood that the present invention is described through some embodiments. Those skilled in the art know that without departing from the spirit and scope of the present invention, various changes or equivalent replacements can be made to these features and embodiments. In addition, under the teaching of the present invention, these features and embodiments can be modified to adapt to specific situations and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application belong to the scope protected by the present invention.
Claims
1. A method for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology, characterized in that The steps are as follows: Obtain MRI data for constructing a pelvic organ model; Perform pelvic organ tissue segmentation based on the MRI data to obtain three-dimensional point cloud data of pelvic organ tissues; wherein, the three-dimensional point cloud data of pelvic organ tissues includes three-dimensional point cloud data of the bladder, three-dimensional point cloud data of the uterus, three-dimensional point cloud data of the endometrium, three-dimensional point cloud data of the vagina, and three-dimensional point cloud data of the rectum; Perform three-dimensional modeling based on the three-dimensional point cloud data of pelvic organ tissues to obtain a solid model of pelvic organ tissues; wherein, the solid model of pelvic organ tissues includes a solid model of the bladder, a solid model of the uterus, a solid model of the endometrium, a solid model of the vagina, and a solid model of the rectum; Construct a bladder cavity model, a muscular uterine cavity model, a vaginal cavity model, and a rectal cavity model based on the solid model of pelvic organ tissues; Perform origin assembly on the bladder cavity model, the muscular uterine cavity model, the vaginal cavity model, and the rectal cavity model to obtain a pelvic organ model.
2. The pelvic organ model construction method based on MRI three-dimensional reconstruction technology according to claim 1, characterized in that, The step of performing pelvic organ tissue segmentation based on the MRI data to obtain three-dimensional point cloud data of pelvic organ tissues specifically includes the following sub-steps: Based on the distribution characteristics of the MRI image gray histogram, respectively outline the initial contours of the bladder, uterus, endometrium, vagina, and rectum, and strengthen the density difference boundaries between the bladder wall, uterine myometrium, endometrium, vaginal wall, and rectal wall and the surrounding tissues to obtain a bladder mask, a uterine mask, an endometrial mask, a vaginal mask, and a rectal mask containing the complete anatomical features of the organs; Perform threshold segmentation, region growing, interlayer contour correction, and mask smoothing processing based on the bladder mask, uterine mask, endometrial mask, vaginal mask, and rectal mask to obtain three-dimensional point cloud data of the bladder, three-dimensional point cloud data of the uterus, three-dimensional point cloud data of the endometrium, three-dimensional point cloud data of the vagina, and three-dimensional point cloud data of the rectum.
3. The method for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology according to claim 2, characterized in that, In the step of respectively outlining the initial contours of the bladder, uterus, endometrium, vagina, and rectum based on the distribution characteristics of the MRI image gray histogram and strengthening the density difference boundaries between the bladder wall, uterine myometrium, endometrium, vaginal wall, and rectal wall and the surrounding tissues, the following steps are further included: Extract wall thickness quantitative parameters for the bladder wall, vaginal wall, and rectal wall respectively, and calculate the average wall thickness of the bladder wall, vaginal wall, and rectal wall respectively to obtain the average bladder wall thickness, the average vaginal wall thickness, and the average rectal wall thickness.
4. The pelvic organ model construction method based on MRI three-dimensional reconstruction technology according to claim 1, wherein The step of performing three-dimensional modeling based on the three-dimensional point cloud data of pelvic organ tissues to obtain a solid model of pelvic organ tissues specifically includes the following sub-steps: Perform nail elimination, point cloud relaxation, noise filtering, hole repair, and surface smoothing processing on the three-dimensional point cloud data of pelvic organ tissues to form a triangular mesh model of pelvic organ tissues; wherein, the triangular mesh model of pelvic organ tissues includes a triangular mesh model of the bladder, a triangular mesh model of the uterus, a triangular mesh model of the uterus, a triangular mesh model of the vagina, and a triangular mesh model of the rectum; Perform non-uniform rational B-spline surface reconstruction based on the triangular mesh model of pelvic organ tissues to obtain a solid model of pelvic organ tissues.
5. The method for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology according to claim 3, wherein In the step of constructing the bladder cavity model, the muscular uterine cavity model, the vaginal cavity model, and the rectal cavity model based on the pelvic organ tissue solid model, the steps of constructing the bladder cavity model are as follows: Perform plane cutting on the bladder solid model, and remove the distal end of the urethral orifice in the bladder solid model to obtain a bladder solid cutting model; Perform shelling on the bladder solid cutting model based on the average bladder wall thickness to obtain an initial bladder cavity model with an open distal end of the urethral orifice and a semi-closed bladder body; Perform nail elimination, mesh relaxation, noise filtering, surface smoothing, non-uniform rational B-spline surface reconstruction, grid construction, and geometric verification on the distal end of the urethral orifice in the initial bladder cavity model to obtain the bladder cavity model.
6. The method for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology according to claim 3, wherein, In the step of constructing the bladder cavity model, the muscular uterine cavity model, the vaginal cavity model, and the rectal cavity model based on the pelvic organ tissue solid model, the steps of constructing the muscular uterine cavity model are as follows: Perform origin assembly on the uterine solid model and the endometrial solid model to obtain an initial muscular uterine solid model; Construct a first model copy of the initial muscular uterine solid model based on the initial muscular uterine solid model to obtain the first model copy of the initial muscular uterine solid model; Perform a Boolean subtraction operation on the initial muscular uterine solid model and the first model copy of the initial muscular uterine solid model to obtain a muscular uterine cavity model with the anatomical morphology of the myometrium.
7. The method for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology according to claim 6, wherein In the step of constructing the bladder cavity model, the muscular uterine cavity model, the vaginal cavity model, and the rectal cavity model based on the pelvic organ tissue solid model, the steps of constructing the vaginal cavity model are as follows: Perform plane cutting on the vaginal solid model, and remove the distal end of the lower segment of the vaginal solid model to obtain a vaginal solid cutting model; Perform shelling on the vaginal solid cutting model based on the average vaginal wall thickness to obtain a first initial vaginal cavity model with a fully closed upper segment and an open lower segment; Perform nail elimination, mesh relaxation, noise filtering, surface smoothing, non-uniform rational B-spline surface reconstruction, grid construction, and geometric verification on the distal end of the lower segment in the first initial vaginal cavity model to obtain a second initial vaginal cavity model; Perform origin assembly on the initial muscular uterine solid model and the second initial vaginal cavity model to obtain a third initial vaginal cavity model; Construct a second model copy of the initial muscular uterine solid model based on the initial muscular uterine solid model to obtain the second model copy of the initial muscular uterine solid model; Perform a Boolean subtraction operation on the third initial vaginal cavity model and the second model copy of the initial muscular uterine solid model to obtain a vaginal cavity model with a cervical anatomical surrounding structure in the upper segment and an open lower segment.
8. The method for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology according to claim 3, characterized in that, In the step of constructing the bladder cavity model, the muscular uterine cavity model, the vaginal cavity model, and the rectal cavity model based on the pelvic organ tissue solid model, the steps of constructing the rectal cavity model are as follows: Perform plane cutting processing based on the rectal solid model, and excise the starting end and the ending end of the rectal solid model to obtain a rectal solid cutting model; Perform shelling processing on the rectal solid cutting model based on the average rectal wall thickness to obtain an initial rectal cavity model with open starting and ending ends; Perform nail elimination, mesh relaxation, noise filtering, surface smoothing, non-uniform rational B-spline surface reconstruction, grid construction, and geometric verification processing on the starting end and the ending end in the initial rectal cavity model respectively to obtain a rectal cavity model.
9. A pelvic organ model construction system based on MRI three-dimensional reconstruction technology, characterized in that It includes an MRI data acquisition module, a pelvic organ tissue segmentation module, a pelvic organ tissue three-dimensional modeling module, a pelvic organ tissue cavity construction module, and a pelvic organ assembly module; wherein, The MRI data acquisition module is used to acquire MRI data for constructing a pelvic organ model; The pelvic organ tissue segmentation module is used to perform pelvic organ tissue segmentation based on the MRI data to obtain three-dimensional point cloud data of pelvic organ tissues; wherein, the three-dimensional point cloud data of pelvic organ tissues includes three-dimensional point cloud data of the bladder, three-dimensional point cloud data of the uterus, three-dimensional point cloud data of the endometrium, three-dimensional point cloud data of the vagina, and three-dimensional point cloud data of the rectum; The pelvic organ tissue three-dimensional modeling module is used to perform three-dimensional modeling based on the three-dimensional point cloud data of pelvic organ tissues to obtain a pelvic organ tissue solid model; wherein, the pelvic organ tissue solid model includes a bladder solid model, a uterine solid model, an endometrial solid model, a vaginal solid model, and a rectal solid model; The pelvic organ tissue cavity construction module is used to construct a bladder cavity model, a muscular uterine cavity model, a vaginal cavity model, and a rectal cavity model based on the pelvic organ tissue solid model; The pelvic organ assembly module is used to perform origin assembly on the bladder cavity model, the muscular uterine cavity model, the vaginal cavity model, and the rectal cavity model to obtain a pelvic organ model.
10. A computer storage medium, characterized in that, The computer storage medium stores a computer program, and the computer program includes program instructions, and when the program instructions are executed by a processor, the processor is caused to execute the method for constructing a pelvic organ model based on MRI three-dimensional reconstruction technology according to any one of claims 1 to 8.