Method, device and equipment for constructing personalized digital human model and storage medium

Abstract

This application discloses a method, apparatus, device, and storage medium for constructing a personalized digital human body model, relating to the field of virtual digital human technology. The method includes: obtaining the age, gender, height, and body mass index (BMI) of a target object; generating a personalized three-dimensional geometric model of the target object based on the age, gender, height, and BMI; calculating biomechanical scaling factors for six regions—the brain, cervical spine, thoracic cavity, abdomen, pelvis, and lower limbs—based on the characteristic parameters of the personalized three-dimensional geometric model and the idealized physical model of each region; and generating a personalized digital human body model based on the biomechanical scaling factors, material properties, and the personalized three-dimensional geometric model. This method can improve the accuracy of constructing personalized digital human body models.

Description

Technical Field

[0001] This application relates to the field of virtual digital human technology, and in particular to a method, apparatus, device and storage medium for constructing a personalized digital human body model. Background Technology

[0002] In the fields of automotive collision safety engineering and biomechanical simulation, constructing high-precision personalized human biomechanical models is a crucial foundation for optimizing vehicle restraint system design, predicting occupant injury, and virtually verifying safety regulations. With increasingly stringent requirements for equitable occupant protection, the rapid generation of personalized human models that accurately represent different ages, genders, heights, and body types (especially non-standard body types such as children, the elderly, and obese individuals) has become a common challenge hindering the development of vehicle passive safety technologies.

[0003] Currently, the mainstream methods in the industry for generating human body models of different body types rely on geometric scaling of a standard percentile human baseline model (usually a 50th percentile male model). Existing technical solutions are mainly divided into two categories: one is global linear geometric scaling, which enlarges or reduces the entire model proportionally based on individual key dimensions (such as height and sitting height); the other is segmental or local geometric scaling, which adjusts the proportions of different segments of the human body (such as the trunk and limbs) according to empirical formulas.

[0004] However, existing technologies have low accuracy when building personalized digital human body models. Summary of the Invention

[0005] This application provides a method, apparatus, device, and storage medium for constructing personalized digital human body models, which can improve the accuracy of constructing personalized digital human body models.

[0006] To achieve the above objectives, this application adopts the following technical solution:

[0007] Firstly, this application provides a method for constructing a personalized digital human body model, including:

[0008] Obtain the target's age, gender, height, and body mass index;

[0009] Based on the age, gender, height, and body mass index, generate a personalized 3D geometric model of the target object;

[0010] For six regions—brain, cervical spine, chest cavity, abdomen, pelvis, and lower limbs—the biomechanical scaling factor for each region is calculated based on the characteristic parameters of the personalized three-dimensional geometric model and the idealized physical model of each region.

[0011] A personalized digital human body model is generated based on the biomechanical scaling factor, material properties, and the personalized three-dimensional geometric model.

[0012] Optionally, the idealized physical model includes a lumped parameter dynamic model and a nonlinear beam. Joint complex model, anisotropic shell element integrated model, viscoelastic fluid Solid coupling model, rigid body Interface soft tissue model, and composite hinge Ligament constraint model.

[0013] Optionally, the biomechanical scaling factors for each region include the acceleration scaling factor and time scaling factor for the cranial region; the axial force scaling factor and bending moment scaling factor for the cervical spine region; the overall stiffness scaling factor for the thoracic region; the overall compressive force scaling factor for the abdominal region; the interface soft tissue stiffness scaling factor for the pelvic region; and the knee joint moment scaling factor for the lower limb region.

[0014] Optionally, the step of calculating the biomechanical scaling factor for each region based on the feature parameters of the personalized three-dimensional geometric model and the idealized physical model of each region includes:

[0015] Feature parameters of the cranial region are extracted based on a personalized 3D geometric model and input into a lumped parameter dynamics model to calculate the acceleration scaling factor and time scaling factor of the cranial region.

[0016] Feature parameters of the cervical spine region are extracted based on a personalized 3D geometric model, and then input into a nonlinear beam. In the joint complex model, calculate the axial force scaling factor and bending moment scaling factor of the cervical spine region;

[0017] Feature parameters of the thoracic cavity region are extracted based on a personalized 3D geometric model and input into an anisotropic shell element integrated model to calculate the overall stiffness scaling factor of the thoracic cavity region.

[0018] Feature parameters of the abdominal region were extracted based on a personalized 3D geometric model, and a viscoelastic fluid was input. In the solid coupling model, calculate the overall compressive force scaling factor for the abdominal region;

[0019] Feature parameters of the pelvic region are extracted based on a personalized 3D geometric model and input into a rigid body. In the interface soft tissue model, calculate the interface soft tissue stiffness scaling factor for the pelvic region;

[0020] Feature parameters of the lower limb region are extracted based on a personalized 3D geometric model, and the composite hinge is input. In the ligament constraint model, calculate the knee joint torque scaling factor in the lower limb region.

[0021] Optionally, the material properties are obtained in the following ways:

[0022] Based on the target object's age and body mass index, the original material properties of the personalized three-dimensional geometric model are adjusted to obtain the material properties.

[0023] Optionally, generating a personalized digital human body model based on the biomechanical scaling factor, material properties, and the personalized three-dimensional geometric model includes:

[0024] Obtain the baseline finite element mesh;

[0025] Based on the personalized three-dimensional geometric model, the baseline finite element mesh is deformed into a mesh of matching shape;

[0026] The mechanical properties of the mesh matching the shape are calibrated according to the biomechanical scaling factor.

[0027] Based on the material properties, material parameters are assigned to the mesh for calibrating mechanical properties to obtain a personalized digital human body model.

[0028] Optionally, the method further includes:

[0029] Based on the target object's body parameters and vehicle interior parameters, a pre-trained statistical model is used to predict the target object's natural posture in the driving environment.

[0030] The personalized digital human body model is adjusted based on the predicted natural posture to generate a personalized digital human body model in a typical driving posture.

[0031] Secondly, this application provides an apparatus for constructing a personalized digital human body model, comprising:

[0032] The acquisition module is used to obtain the target object's age, gender, height, and body mass index.

[0033] The processing module is used to generate a personalized three-dimensional geometric model of the target object based on the age, gender, height and body mass index; and to calculate the biomechanical scaling factor of each of the six regions—brain, cervical spine, chest cavity, abdomen, pelvis and lower limbs—based on the feature parameters of the personalized three-dimensional geometric model and the idealized physical model of each region.

[0034] The generation module is used to generate a personalized digital human body model based on the biomechanical scaling factor, material properties, and the personalized three-dimensional geometric model.

[0035] Thirdly, this application provides a computing device, including a memory and a processor;

[0036] The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of the first aspects.

[0037] Fourthly, this application provides a computer-readable storage medium for storing a computer program for performing the method as described in any one of the first aspects.

[0038] As can be seen from the above technical solution, this application has at least the following beneficial effects:

[0039] In this application, a personalized three-dimensional geometric model is generated by combining the target object's age, gender, height, and body mass index, which accurately captures the nonlinear anatomical morphological characteristics of different groups, such as the head-to-body ratio of children, kyphosis of the elderly, and trunk shape of obese people, thus solving the problem of geometric distortion in traditional methods from the source.

[0040] Dedicated idealized physical models were constructed for six regions: cranium, cervical spine, thoracic cavity, abdomen, pelvis, and lower limbs. Based on personalized geometric feature parameters, targeted biomechanical scaling factors were calculated. At the same time, material properties were adaptively adjusted in combination with age and body mass index to ensure that the model not only fits the target object in shape, but also is highly consistent with the target individual in mechanical response characteristics (such as acceleration, stiffness, and torque). This significantly improves the accuracy of injury prediction for non-standard body types and effectively supports the equitable protection design of vehicle restraint systems.

[0041] The entire methodology is logically closed-loop and can be efficiently implemented. From parameter acquisition to model generation, no complex manual intervention is required. Furthermore, the pre-trained statistical model can be used to further achieve accurate matching and positioning of driving posture. The generated model can be directly used for vehicle constraint system optimization, occupant injury prediction, and virtual verification of safety regulations, providing tool support for the upgrading of vehicle passive safety technology and reducing the R&D cost and cycle of safety design.

[0042] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0043] Figure 1 A flowchart illustrating a method for constructing a personalized digital human body model, as provided in this application embodiment;

[0044] Figure 2 A schematic diagram of an apparatus for constructing a personalized digital human body model provided in an embodiment of this application;

[0045] Figure 3 This is a schematic diagram of a computing device provided in an embodiment of this application. Detailed Implementation

[0046] The terms "first," "second," and "third," etc., used in this application specification and accompanying drawings are used to distinguish different objects, not to limit a specific order.

[0047] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0048] To ensure clarity and conciseness in the description of the following embodiments, a brief introduction to the related technologies is given first:

[0049] Personalized digital human models are digital models constructed based on physiological parameters such as age, gender, height, and body mass index of the target object. They can accurately reflect an individual's anatomical morphology, material properties, and biomechanical response, and are tools for automobile collision safety simulation and constraint system design optimization.

[0050] In the fields of automotive collision safety engineering and biomechanical simulation, high-precision personalized human body models are the foundation for optimizing vehicle restraint systems, predicting occupant injuries, and virtually verifying safety regulations. With the increasing demands for equitable occupant protection, traditional model building methods are no longer sufficient to meet practical needs, and key technical problems are becoming increasingly prominent: the accuracy of personalized human body models generated by existing technologies is low, especially for non-standard body types such as children, the elderly, and obese individuals, where the prediction error for biomechanical responses is large. This fails to provide reliable support for equitable vehicle safety design and has become a common bottleneck restricting the upgrading of passive safety technologies.

[0051] The root cause of this problem lies in the inherent flaws of traditional methods that rely primarily on geometric scaling. On the one hand, geometric scaling is based on the assumption that different parts of the human body change proportionally, ignoring the nonlinear characteristics of human anatomy and failing to reproduce key features such as differences in head-to-body ratio in children, kyphosis in the elderly, and the unique trunk shape of obese individuals, resulting in distorted anatomical morphology of the model. On the other hand, traditional methods assume that material properties remain constant, failing to consider age-related decreases in bone density, changes in soft tissue viscoelasticity, and the imbalance in fat-to-muscle ratio caused by BMI. Furthermore, they lack consideration of the specific biomechanical mechanisms of different regions, and simply adjusting the size cannot guarantee dynamic similarity at the kinetic level, ultimately leading to a significant deviation between the model's mechanical response and that of the target individual.

[0052] In view of this, embodiments of this application provide a method for constructing a personalized digital human body model, which can be executed by a processing device. The processing device can be a terminal or a server. Terminals include, but are not limited to, in-vehicle terminals, smartphones, tablets, laptops, personal digital assistants, or smart wearable devices. The server can be a cloud server, such as a central server in a central cloud computing cluster or an edge server in an edge cloud computing cluster. Alternatively, the server can be a server in a local data center. A local data center refers to a data center directly controlled by the user.

[0053] Addressing the core issues of low accuracy and distorted mechanical response prediction in existing technologies that rely on geometric scaling for non-standard body type models, this application takes the target subject's age, gender, height, and body mass index as input. First, it overcomes the limitations of traditional geometric scaling to generate a personalized 3D geometric model that accurately reflects individual anatomical characteristics. Then, for six key biomechanical regions of the human body, including the brain and cervical spine, it uses a dedicated idealized physical model to uncover the intrinsic correlation between geometric features and mechanical response, deriving region-specific biomechanical scaling factors. Simultaneously, it adaptively adjusts material properties based on age and body mass index to overcome the rigidity of material properties in traditional methods. Finally, through an integrated process of mesh deformation, mechanical calibration, and material assignment, it deeply integrates geometric morphology, mechanical properties, and material attributes to construct a personalized digital human body model that combines anatomical realism and mechanical fidelity. Furthermore, it can accurately adapt the model to application scenarios through driving posture prediction, fundamentally solving the technical bottleneck of large prediction errors in non-standard body type models and supporting equal safety design for vehicles.

[0054] To make the technical solution of this application clearer and easier to understand, the method for constructing a personalized digital human body model provided by the embodiments of this application will be described below with reference to the accompanying drawings. Figure 1 As shown, this figure is a flowchart of a method for constructing a personalized digital human body model according to an embodiment of this application. The method includes:

[0055] S201, The processing device acquires the age, gender, height, and body mass index of the target object.

[0056] The target group refers to specific individuals who need to build personalized digital human body models. These individuals can be of any age (such as children, adults, and the elderly), gender, height, and body mass index, especially non-standard body types such as children, the elderly, and obese individuals. This group is the target group for this application to address the problem of model construction accuracy.

[0057] Age refers to the actual physiological age of the target object, and is a physiological parameter that affects human anatomical morphology (such as bone development and spinal morphology) and material properties (such as bone density and soft tissue elasticity).

[0058] Gender refers to the biological sex (male / female) of the target object. There are significant differences in the human body shape (such as shoulder width, pelvic width, muscle distribution) and biomechanical characteristics (such as bone stiffness, soft tissue viscoelasticity) of individuals of different genders, which are important distinguishing parameters for the construction of personalized models.

[0059] Height refers to the net height of the target object, that is, the vertical distance from the top of the head to the bottom of the feet when standing. It is a parameter that reflects the overall shape and scale of the human body and directly affects the overall size benchmark of the personalized three-dimensional geometric model.

[0060] Body Mass Index (BMI) is a value calculated by dividing weight (kg) by height (m)². It is a key indicator for measuring the degree of fatness and body type of a person, such as underweight, normal weight, overweight, and obese. It can directly reflect the ratio of fat to muscle tissue in the target object and is the basis for adjusting the material properties of the model and optimizing the trunk shape.

[0061] Specifically, the processing equipment collects four physiological and physical parameters of the specific individual (target object) whose personalized model needs to be built, through preset data acquisition methods, including manual input, data transmission from intelligent measurement devices, and targeted retrieval from related databases. These parameters are age, gender, height, and body mass index (BMI). These four parameters are the basic inputs driving all subsequent technical processes, each playing an indispensable role. Age and gender together determine the basic anatomical morphological characteristics of the human body and also govern the laws of physiological development and decline, such as bone growth in childhood and decreased bone density in old age. Height provides an overall dimensional benchmark for the personalized 3D geometric model, directly relating to the overall size range of the model. BMI directly reflects the target object's body fatness or leanness, and can then be used to deduce the proportion of fat and muscle tissue in their body, providing a basis for subsequent material property adjustments.

[0062] S202. The processing device generates a personalized three-dimensional geometric model of the target object based on age, gender, height, and body mass index.

[0063] Personalized 3D geometric models refer to 3D digital models constructed based on the specific physiological and physical parameters of the target object, which can reflect the anatomical structure and morphological characteristics of that individual. They are different from the industry-standard percentile human body models and can reflect the specific morphology of people of different ages, genders, and body types, such as the large head-to-body ratio of children, the kyphosis of the elderly, and the trunk outline of obese people.

[0064] The processing device generates a personalized 3D geometric model of the target object based on age, gender, height, and body mass index. This is the step in this application from inputting basic parameters to realizing personalized model morphology, and it is also a key link to break through the limitations of traditional global or segmented geometric scaling.

[0065] Specifically, the processing device's execution logic is as follows: First, it calls a pre-defined large-scale multimodal human morphology database (including optical scan point clouds and CT / MRI medical image data). After preprocessing the data (Protocol analysis to eliminate translation / rotation / scaling differences, and homeomorphic mesh deformation to achieve topological uniformity), it extracts K principal components (PCA) that cover the vast majority of morphological variations and determines the mean shape of all samples. The purpose of this step is to transform the complex human morphology into a concise parametric form of a weighted combination of the mean shape and principal components; that is, the morphology of any individual can be represented by the mean shape superimposed with principal components of different weights. The calculation expression is:

[0066]

[0067] in, Represents the three-dimensional geometric shape of any individual. This represents the shape of the mean of all samples in the database. Indicates the first The weights of each principal component, Indicates the first Principal components, Indicates the number of principal components.

[0068] Subsequently, the processing device calls a trained nonlinear machine learning model, such as Gaussian process regression, which has established a mapping relationship between age, gender, height, BMI and principal component weight vectors through database samples.

[0069] By inputting the four parameters of the target object into this nonlinear machine learning model, a unique shape weight vector can be predicted. Substituting this vector into the parameterization formula yields the initial 3D geometric model. The calculation expression is:

[0070]

[0071] in, Represents the initial three-dimensional geometric model. This represents the initial morphological weight vector predicted by the machine learning model.

[0072] Finally, to improve the fidelity of key anatomical regions, the processing device introduces anatomical landmark constraints (such as vertebral spinous processes and rib arch apexes). By constructing a loss function and optimizing the solution, an adjusted optimal weight vector is obtained, ultimately generating a personalized 3D geometric model that accurately matches the anatomical features of the target object. The expression is:

[0073]

[0074] in, Represents a personalized 3D geometric model. This represents the adjusted optimal weight vector.

[0075] S203 The processing device targets six regions: the brain, cervical spine, chest cavity, abdomen, pelvis, and lower limbs. Based on the characteristic parameters of the personalized three-dimensional geometric model and the idealized physical model of each region, it calculates the biomechanical scaling factor for each region.

[0076] Idealized physical models include lumped parameter dynamic models and nonlinear beams. Joint complex model, anisotropic shell element integrated model, viscoelastic fluid Solid coupling model, rigid body Interface soft tissue model, and composite hinge Ligament constraint model.

[0077] The biomechanical scaling factors for each region include the acceleration scaling factor and time scaling factor for the cranial region; the axial force scaling factor and bending moment scaling factor for the cervical spine region; the overall stiffness scaling factor for the thoracic region; the overall compressive force scaling factor for the abdominal region; the interface soft tissue stiffness scaling factor for the pelvic region; and the knee joint moment scaling factor for the lower limb region.

[0078] Specifically, the processing device extracts feature parameters of the cranial region based on a personalized three-dimensional geometric model and inputs them into a lumped parameter dynamics model to calculate the acceleration scaling factor and time scaling factor of the cranial region.

[0079] Parameters such as mass and feature dimensions of the cranium are extracted from a personalized 3D geometric model and input into a lumped parameter dynamics model to calculate the acceleration scaling factor and time scaling factor of the cranium region. The calculation expressions are as follows:

[0080]

[0081]

[0082] in, The acceleration scaling factor for the cranial region is used to calibrate the model's cranial acceleration response. The time scaling factor representing the brain region is used to calibrate the brain dynamic response time characteristics of the model. The scaling factor for the shear modulus of the cranial soft tissue is derived from the cranial dimensions of the personalized geometric model combined with the material parameters related to the age of the target object. The scaling factor for brain tissue density is determined by the brain volume of the personalized geometric model. Derivation of mass ratio; This represents the cranial geometry scaling factor, which is the ratio of the cranial feature dimensions of the personalized geometric model to those of the baseline model.

[0083] Feature parameters of the cervical spine region are extracted based on a personalized 3D geometric model, and then input into a nonlinear beam. In the joint complex model, calculate the axial force scaling factor and bending moment scaling factor of the cervical spine region;

[0084] Parameters such as cross-sectional dimensions and segment lengths of the cervical spine are extracted from a personalized 3D geometric model and input into a nonlinear beam. Using a joint complex model, calculate the scaling factors for axial force and bending moment in the cervical spine region. The calculation expressions are as follows:

[0085]

[0086]

[0087] in, This represents the axial force scaling factor for the cervical spine region, used to calibrate the cervical spine axial load response of the model. This represents the scaling factor for the bending moment in the cervical spine region, used to calibrate the model's cervical spine bending load response. This represents the scaling factor for the elastic modulus of the cervical spine, derived from the cervical spine bone density correlation parameters of the personalized geometric model. The geometric scaling factor, representing the contact area between the cervical cortex and bone, is obtained from measurements of the cervical spine cross-sectional dimensions of a personalized 3D geometric model. This represents the scaling factor for the stiffness of the cervical ligaments. The geometric scaling factor representing the thickness of the cervical intervertebral disc is obtained from the cervical segment length measured from a personalized 3D geometric model.

[0088] Feature parameters of the thoracic cavity region are extracted based on a personalized 3D geometric model and input into an anisotropic shell element integrated model to calculate the overall stiffness scaling factor of the thoracic cavity region.

[0089] Parameters such as rib curvature and thoracic dimensions of the thoracic cavity are extracted from a personalized 3D geometric model and input into an anisotropic shell element integrated model to calculate the overall stiffness scaling factor of the thoracic cavity region. The calculation expression is as follows:

[0090]

[0091] in, This represents the overall stiffness scaling factor for the thoracic region, used to calibrate the thoracic deformation stiffness of the model. This represents the scaling factor for the elastic modulus of the ribs in the thoracic cavity, which is derived from the parameters related to the cross-sectional dimensions of the ribs in the personalized geometric model. This represents the scaling factor for rib thickness in the thoracic cavity, which is the ratio of the rib thickness in the personalized geometry model to that in the baseline model. This represents the thoracic geometry scaling factor, which is the ratio of the thoracic diameter of the personalized geometry model to that of the baseline model.

[0092] Feature parameters of the abdominal region were extracted based on a personalized 3D geometric model, and a viscoelastic fluid was input. In the solid coupling model, calculate the overall compressive force scaling factor for the abdominal region;

[0093] Parameters such as the volume and wall thickness of the abdomen are extracted from a personalized 3D geometric model, and then a viscoelastic fluid is input. A solid-coupled model is used to calculate the global compressive force scaling factor in the abdominal region. The calculation expression is as follows:

[0094]

[0095]

[0096]

[0097] in, This represents the overall compressive force scaling factor for the abdominal region, used to calibrate the total compressive load response of the abdomen in obese individuals. This represents the first weighting coefficient related to BMI. This represents the second weighting coefficient related to BMI. The compressibility factor of the abdominal viscoelastic solid (organ). The elastic modulus scaling factor for abdominal organs is derived from organ material properties associated with BMI (such as changes in organ elasticity in obese individuals). The geometric scaling factor, representing the organ contact area, is calculated from the organ dimensions extracted from the personalized 3D geometric model. The compressibility factor representing the incompressible fluid in the abdomen (abdominal contents). The dynamic viscosity scaling factor for peritoneal fluid is derived from the composition of peritoneal contents (such as fat percentage) associated with BMI. The surface tension scaling factor for peritoneal fluid is derived from the peritoneal environment characteristics associated with BMI. The geometric scaling factor representing the inter-organ spacing is calculated from the inter-organ spacing extracted from the personalized 3D geometric model.

[0098] Feature parameters of the pelvic region are extracted based on a personalized 3D geometric model and input into a rigid body. In the interface soft tissue model, calculate the interface soft tissue stiffness scaling factor for the pelvic region;

[0099] Parameters such as interosseous distance and soft tissue thickness of the pelvis are extracted from a personalized 3D geometric model and input into a rigid body. The interface soft tissue model is used to calculate the stiffness scaling factor of the interface soft tissue in the pelvic region. The calculation expression is:

[0100]

[0101] in, This represents a scaling factor for the interface soft tissue stiffness of the pelvic region, used to calibrate the pelvis of the model. Soft tissue interface stiffness; This represents the scaling factor for the shear modulus of the soft tissue around the pelvis, which is derived from the soft tissue thickness-related parameters of the personalized geometric model. This represents the ratio of the soft tissue thickness of the personalized geometric model to that of the baseline model, indicating the scaling factor for the soft tissue thickness around the pelvis. This represents the ratio of the interosseous distance of the pelvis in the personalized geometric model to that in the baseline model, indicating the pelvic geometry scaling factor.

[0102] Feature parameters of the lower limb region are extracted based on a personalized 3D geometric model, and the composite hinge is input. In the ligament constraint model, the knee joint moment scaling factor in the lower limb region is calculated. Parameters such as ligament length and joint space of the knee joint are extracted from a personalized 3D geometric model and input as a composite hinge. Using a ligament-constrained model, calculate the knee joint moment scaling factor in the lower limb region. The calculation expression is:

[0103]

[0104] in, This represents the knee joint torque scaling factor for the lower limb region, used to calibrate the knee joint torque response of the model. This represents the knee ligament stiffness scaling factor, derived from the ligament cross-sectional dimensions associated parameters of the personalized geometric model; This represents the lower limb geometric scaling factor, which is the ratio of the knee ligament length of the personalized geometric model to that of the baseline model.

[0105] S204. The processing equipment generates a personalized digital human body model based on the biomechanical scaling factor, material properties, and personalized three-dimensional geometric model.

[0106] Material properties are obtained by adjusting the original material properties of the personalized 3D geometric model based on the target object's age and body mass index.

[0107] Material properties refer to the mechanical characteristics of human tissues (such as muscles, fat, bones, organs, etc.), including elastic modulus, shear modulus, density, viscosity, etc. These parameters determine the mechanical response of tissues under stress, such as deformation and load-bearing capacity, and are the basis for achieving accurate mechanical simulation of personalized digital human body models.

[0108] Original material properties refer to the standard material parameters initially associated with a personalized 3D geometric model. These are usually based on the average tissue characteristics of a general human database or the default material parameters of a baseline model, and are not adjusted for the individual characteristics of the target object.

[0109] Material property adjustment refers to the process of modifying the original material properties based on the specific physiological parameters of the target object (such as age and BMI) so that the final material properties conform to the actual tissue mechanical properties of the individual.

[0110] Specifically: After a personalized 3D geometric model is generated, it will be associated with a set of original material properties by default (based on the average tissue characteristics of a general database); however, the mechanical properties of tissues vary significantly among individuals of different ages and BMIs. For example, the elastic modulus of bones is lower in the elderly and the shear modulus of fat is lower in obese individuals. Therefore, the processing device will call up preset material properties based on the target object's age and BMI. Physiological parameter correlation models are used to make targeted corrections to the original material properties, ultimately resulting in exclusive material properties that fit the individual's actual situation.

[0111] For bone tissue, soft tissue, and tissue components, the processing device applies different adjustment rules based on the target's age and BMI. For bone tissue, a piecewise function is used to represent the decay of Young's modulus with age, calculated as follows:

[0112]

[0113] in, This represents the Young's modulus of cortical bone when the target subject is age A. This indicates a reference to Young's modulus, for example, for women. =17GPa; This indicates the age corresponding to peak bone mineral density, for example, in women. =30; Indicates the first attenuation coefficient. This represents the second attenuation coefficient.

[0114] For soft tissue, the changes in age / BMI are correlated with the parameters of the viscoelastic model (relaxation time, equilibrium modulus), and the expression is:

[0115]

[0116] in, This represents the soft tissue relaxation time constant when the target subject is of age A; Indicates the reference relaxation time constant. Represents the age correlation coefficient. Indicates a reference age.

[0117] For tissue composition, the proportion of fat is estimated using BMI, and then the volume fraction of fat / muscle and material parameters are adjusted. Ultimately, this ensures that the material properties of the finite element model closely match the actual physiological state of the target object, avoiding simulation deviations caused by generic parameters.

[0118] The specific steps for generating a personalized digital human body model are as follows:

[0119] First, the processing device acquires a baseline finite element mesh; based on the personalized 3D geometric model, the baseline finite element mesh is deformed into a mesh with a matching shape; based on the biomechanical scaling factor, the mechanical properties of the mesh with the matching shape are calibrated; based on the material properties, material parameters are assigned to the mesh with calibrated mechanical properties to obtain a personalized digital human body model.

[0120] The processing equipment first retrieves the industry-standard baseline finite element mesh, a standardized human body structure mesh (including nodes and element connections), which serves as the basic template for generating personalized models. Based on the previously generated personalized 3D geometric model (the exclusive shape of the target object), the processing equipment deforms and adjusts the baseline finite element mesh to ensure that the shape and size of the mesh accurately match the anatomical shape of the target object. For example, it deforms the mesh of a standard body type into the proportions of a child's head and body, or the torso outline of an obese person.

[0121] Combining the previously calculated biomechanical scaling factor (mechanical correction coefficient for areas such as the brain, cervical spine, and abdomen), the processing device adjusts the mechanical parameters of the deformed mesh to match the mesh's mechanical response (such as acceleration, stiffness, and load-bearing capacity) with the actual physiological characteristics of the target object. For example, it makes the bone stiffness of the elderly model match the state after its bone density has decreased.

[0122] Finally, the processing equipment assigns the previously adjusted target object-specific material properties, such as Young's modulus of bone and viscoelastic parameters of soft tissue, to the calibrated mechanical properties grid, thus completing the personalized assignment of material parameters.

[0123] After these four steps, the original general baseline model is transformed into a personalized digital human body model that has a unique shape, unique mechanical response, and unique material properties, which can be used for accurate biomechanical simulation (such as damage prediction and load analysis).

[0124] The method also includes:

[0125] The processing device predicts the natural posture of the target object in the driving environment through a pre-trained statistical model based on the target object's body parameters and vehicle interior parameters; it then adjusts the personalized digital human body model according to the predicted natural posture to generate a personalized digital human body model in a typical driving posture.

[0126] The processing device first collects the target object's body parameters (such as height, arm length, and leg length) and vehicle interior parameters (such as seat position, steering wheel angle, and pedal spacing). These parameters are then input into a pre-trained statistical model, which is trained based on a large amount of driver posture data. Through the statistical model, the target object's natural posture while driving in the vehicle is calculated, which is a comfortable posture that conforms to its physical condition and driving habits, such as the angle of arm bending, the degree of leg extension, and the tilt of the seat.

[0127] The processing device uses the predicted natural driving posture as instructions to perform posture-driven (adjusting the model's joint angles and limb positions) and positioning adjustments (placing the model in the corresponding position in the vehicle's interior and matching the relative spatial relationship between the seat and steering wheel) on the generated personalized digital human body model. Finally, a personalized digital human body model in a typical driving posture is obtained. At this point, the model retains the shape, mechanics, and material-specific characteristics of the target object, and is in the object's real driving posture, which can be used for subsequent driving comfort analysis, collision safety simulation, and other scenarios.

[0128] Based on the above description, this application has the following beneficial effects:

[0129] In this application, a personalized three-dimensional geometric model is generated by combining the target object's age, gender, height, and body mass index, which accurately captures the nonlinear anatomical morphological characteristics of different groups, such as the head-to-body ratio of children, kyphosis of the elderly, and trunk shape of obese people, thus solving the problem of geometric distortion in traditional methods from the source.

[0130] Dedicated idealized physical models were constructed for six regions: cranium, cervical spine, thoracic cavity, abdomen, pelvis, and lower limbs. Based on personalized geometric feature parameters, targeted biomechanical scaling factors were calculated. At the same time, material properties were adaptively adjusted in combination with age and body mass index to ensure that the model not only fits the target object in shape, but also is highly consistent with the target individual in mechanical response characteristics (such as acceleration, stiffness, and torque). This significantly improves the accuracy of injury prediction for non-standard body types and effectively supports the equitable protection design of vehicle restraint systems.

[0131] The entire methodology is logically closed-loop and can be efficiently implemented. From parameter acquisition to model generation, no complex manual intervention is required. Furthermore, the pre-trained statistical model can be used to further achieve accurate matching and positioning of driving posture. The generated model can be directly used for vehicle constraint system optimization, occupant injury prediction, and virtual verification of safety regulations, providing tool support for the upgrading of vehicle passive safety technology and reducing the R&D cost and cycle of safety design.

[0132] The above text combined Figure 1The method for constructing a personalized digital human body model provided in the embodiments of this application has been described in detail. The apparatus and equipment provided in the embodiments of this application will be described below with reference to the accompanying drawings.

[0133] like Figure 2 As shown in the figure, this is a schematic diagram of an apparatus for constructing a personalized digital human body model according to an embodiment of this application. The apparatus includes:

[0134] The acquisition module 301 is used to acquire the age, gender, height, and body mass index of the target object;

[0135] The processing module 302 is used to generate a personalized three-dimensional geometric model of the target object based on the age, gender, height and body mass index; and to calculate the biomechanical scaling factor of each region for six regions: cranium, cervical spine, chest cavity, abdomen, pelvis and lower limbs, based on the feature parameters of the personalized three-dimensional geometric model and the idealized physical model of each region.

[0136] The generation module 303 is used to generate a personalized digital human body model based on the biomechanical scaling factor, material properties, and the personalized three-dimensional geometric model.

[0137] Optionally, the processing module 302 is specifically used to extract feature parameters of the cranial region based on the personalized three-dimensional geometric model, input them into the lumped parameter dynamic model, and calculate the acceleration scaling factor and time scaling factor of the cranial region.

[0138] Feature parameters of the cervical spine region are extracted based on a personalized 3D geometric model, and then input into a nonlinear beam. In the joint complex model, calculate the axial force scaling factor and bending moment scaling factor of the cervical spine region;

[0139] Feature parameters of the thoracic cavity region are extracted based on a personalized 3D geometric model and input into an anisotropic shell element integrated model to calculate the overall stiffness scaling factor of the thoracic cavity region.

[0140] Feature parameters of the abdominal region were extracted based on a personalized 3D geometric model, and a viscoelastic fluid was input. In the solid coupling model, calculate the overall compressive force scaling factor for the abdominal region;

[0141] Feature parameters of the pelvic region are extracted based on a personalized 3D geometric model and input into a rigid body. In the interface soft tissue model, calculate the interface soft tissue stiffness scaling factor for the pelvic region;

[0142] Feature parameters of the lower limb region are extracted based on a personalized 3D geometric model, and the composite hinge is input. In the ligament constraint model, calculate the knee joint torque scaling factor in the lower limb region.

[0143] Optionally, the processing module 302 is specifically used to adjust the original material properties of the personalized three-dimensional geometric model according to the age and body mass index of the target object, so as to obtain the material properties.

[0144] Optionally, the generation module 303 is specifically used to obtain the baseline finite element mesh;

[0145] Based on the personalized three-dimensional geometric model, the baseline finite element mesh is deformed into a mesh of matching shape;

[0146] The mechanical properties of the mesh matching the shape are calibrated according to the biomechanical scaling factor.

[0147] Based on the material properties, material parameters are assigned to the mesh for calibrating mechanical properties to obtain a personalized digital human body model.

[0148] Optionally, the generation module 303 is also used to predict the natural posture of the target object in the driving environment by using a pre-trained statistical model based on the target object's body parameters and vehicle interior parameters.

[0149] The personalized digital human body model is adjusted based on the predicted natural posture to generate a personalized digital human body model in a typical driving posture.

[0150] The apparatus for constructing a personalized digital human body model according to the embodiments of this application can correspond to performing the method described in the embodiments of this application, and the other operations and / or functions of the various modules / units of the apparatus for constructing a personalized digital human body model are respectively for implementing Figure 1 For the sake of brevity, the corresponding processes of each method in the illustrated embodiments will not be described in detail here.

[0151] This application also provides a computing device. For example... Figure 3 As shown in the figure, this is a schematic diagram of a computing device provided in an embodiment of this application. The computing device 700 includes a bus 701, a processor 702, a communication interface 703, and a memory 704. The processor 702, the memory 704, and the communication interface 703 communicate with each other via the bus 701.

[0152] The 701 bus can be a standard bus for interconnecting peripheral components or an extended industry standard structure bus. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0153] The processor 702 can be any one or more of the following: a central processing unit, a graphics processing unit, a microprocessor, or a digital signal processor.

[0154] The communication interface 703 is used for communication with external devices.

[0155] Memory 704 may include volatile memory, such as random access memory. Memory 704 may also include non-volatile memory, such as read-only memory, flash memory, hard disk drive, or solid-state drive.

[0156] The memory 704 stores executable code, which the processor 702 executes to perform the aforementioned method for constructing a personalized digital human body model.

[0157] Specifically, in achieving Figure 2 In the case of the illustrated embodiment, and Figure 2 When the modules or units of the apparatus for constructing a personalized digital human body model described in the embodiments are implemented by software, the execution... Figure 2 The software or program code required for the functions of each module / unit can be partially or wholly stored in the memory 704. The processor 702 executes the program code corresponding to each unit stored in the memory 704, and executes the aforementioned method for constructing a personalized digital human body model.

[0158] This application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that a computing device can store, or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive). The computer-readable storage medium includes instructions that instruct a computing device to execute the above-described method for constructing a personalized digital human body model.

[0159] This application also provides a computer program product comprising one or more computer instructions. When the computer instructions are loaded and executed on a computing device, all or part of the processes or functions described in this application are generated.

[0160] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0161] When the computer program product is executed by a computer, the computer performs any of the aforementioned methods for constructing a personalized digital human body model. The computer program product can be a software installation package; when any of the aforementioned methods for constructing a personalized digital human body model is required, the computer program product can be downloaded and executed on the computer.

[0162] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.

[0163] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be covered within the scope of protection of this application.

Claims

1. A method for constructing a personalized digital human body model, characterized in that, The method includes: Obtain the target's age, gender, height, and body mass index; Based on the age, gender, height, and body mass index, generate a personalized 3D geometric model of the target object; For six regions—brain, cervical spine, chest cavity, abdomen, pelvis, and lower limbs—the biomechanical scaling factor for each region is calculated based on the characteristic parameters of the personalized three-dimensional geometric model and the idealized physical model of each region. A personalized digital human body model is generated based on the biomechanical scaling factor, material properties, and the personalized three-dimensional geometric model. The material properties are obtained in the following way: Based on the target object's age and body mass index, the original material properties of the personalized three-dimensional geometric model are adjusted to obtain the material properties; The biomechanical scaling factors for each region include the acceleration scaling factor and time scaling factor for the cranial region; the axial force scaling factor and bending moment scaling factor for the cervical spine region; the overall stiffness scaling factor for the thoracic region; the overall compressive force scaling factor for the abdominal region; the interface soft tissue stiffness scaling factor for the pelvic region; and the knee joint moment scaling factor for the lower limb region. The acceleration scaling factor and the time scaling factor are obtained in the following way: in, This represents the acceleration scaling factor for the cranial region. The time scaling factor representing the brain region. This represents the scaling factor for the shear modulus of soft tissue in the brain. This represents the scaling factor for the density of brain tissue. Indicates the geometric scaling factor of the brain; The axial force scaling factor and the bending moment scaling factor are obtained as follows: in, This represents the scaling factor for axial force in the cervical spine region. This represents the scaling factor for the bending moment in the cervical spine region. This represents the scaling factor for the elastic modulus of the cervical vertebrae. The geometric scaling factor representing the contact area between the cervical cortex and bone. This represents the scaling factor for the stiffness of the cervical ligaments. Geometric scaling factor representing the thickness of the cervical intervertebral disc; The overall stiffness scaling factor is obtained as follows: in, This represents the overall stiffness scaling factor for the thoracic cavity region. This represents the scaling factor for the elastic modulus of the ribs in the thoracic cavity. This represents the scaling factor for the thickness of the ribs in the thoracic cavity. Indicates the geometric scaling factor of the thoracic cavity; The overall compressive force scaling factor is obtained as follows: in, This represents the overall compressive force scaling factor for the abdominal region. This represents the first weighting coefficient related to BMI. This represents the second weighting coefficient related to BMI. The compressibility factor of a viscoelastic solid in the abdomen. This represents the scaling factor for the elastic modulus of abdominal organs. The geometric scaling factor representing the contact area of ​​the organ. The compressibility factor represents the compressibility factor of the incompressible fluid in the abdomen. This represents the scaling factor for the dynamic viscosity of the peritoneal fluid. This represents the scaling factor for the surface tension of the peritoneal fluid. The geometric scaling factor representing the inter-organ space; The interface soft tissue stiffness scaling factor is obtained in the following way: in, This represents the scaling factor for the interface soft tissue stiffness in the pelvic region. This represents the scaling factor for the shear modulus of the soft tissues surrounding the pelvis. This represents the ratio of the soft tissue thickness of the personalized geometric model to that of the baseline model, indicating the scaling factor for the soft tissue thickness around the pelvis. This represents the ratio of the interosseous distance of the pelvis in the personalized geometric model to that in the baseline model, indicating the pelvic geometry scaling factor. The knee joint torque scaling factor is obtained in the following way: in, This represents the knee joint torque scaling factor for the lower limb region. This represents the scaling factor for knee ligament stiffness. This represents the geometric scaling factor for the lower limbs.

2. The method according to claim 1, characterized in that, The idealized physical model includes a lumped parameter dynamic model and a nonlinear beam. Joint complex model, anisotropic shell element integrated model, viscoelastic fluid Solid coupling model, rigid body Interface soft tissue model, and composite hinge Ligament constraint model.

3. The method according to claim 1, characterized in that, The step of calculating the biomechanical scaling factor for each region based on the feature parameters of the personalized 3D geometric model and the idealized physical model of each region includes: Feature parameters of the cranial region are extracted based on a personalized 3D geometric model and input into a lumped parameter dynamics model to calculate the acceleration scaling factor and time scaling factor of the cranial region. Feature parameters of the cervical spine region are extracted based on a personalized 3D geometric model, and then input into a nonlinear beam. In the joint complex model, calculate the axial force scaling factor and bending moment scaling factor of the cervical spine region; Feature parameters of the thoracic cavity region are extracted based on a personalized 3D geometric model and input into an anisotropic shell element integrated model to calculate the overall stiffness scaling factor of the thoracic cavity region. Feature parameters of the abdominal region were extracted based on a personalized 3D geometric model, and a viscoelastic fluid was input. In the solid coupling model, calculate the overall compressive force scaling factor for the abdominal region; Feature parameters of the pelvic region are extracted based on a personalized 3D geometric model and input into a rigid body. In the interface soft tissue model, calculate the interface soft tissue stiffness scaling factor for the pelvic region; Feature parameters of the lower limb region are extracted based on a personalized 3D geometric model, and the composite hinge is input. In the ligament constraint model, calculate the knee joint torque scaling factor in the lower limb region.

4. The method according to claim 1, characterized in that, The process of generating a personalized digital human body model based on the biomechanical scaling factor, material properties, and the personalized three-dimensional geometric model includes: Obtain the baseline finite element mesh; Based on the personalized three-dimensional geometric model, the baseline finite element mesh is deformed into a mesh of matching shape; The mechanical properties of the mesh matching the shape are calibrated according to the biomechanical scaling factor. Based on the material properties, material parameters are assigned to the mesh for calibrating mechanical properties to obtain a personalized digital human body model.

5. The method according to claim 1, characterized in that, The method further includes: Based on the target object's body parameters and vehicle interior parameters, a pre-trained statistical model is used to predict the target object's natural posture in the driving environment. The personalized digital human body model is adjusted based on the predicted natural posture to generate a personalized digital human body model in a typical driving posture.

6. A device for constructing a personalized digital human body model, characterized in that, The device includes: The acquisition module is used to obtain the target object's age, gender, height, and body mass index. The processing module is used to generate a personalized three-dimensional geometric model of the target object based on the age, gender, height and body mass index; and to calculate the biomechanical scaling factor of each of the six regions—brain, cervical spine, chest cavity, abdomen, pelvis and lower limbs—based on the feature parameters of the personalized three-dimensional geometric model and the idealized physical model of each region. The generation module is used to generate a personalized digital human body model based on the biomechanical scaling factor, material properties, and the personalized three-dimensional geometric model. The material properties are obtained by adjusting the original material properties of the personalized three-dimensional geometric model according to the target object's age and body mass index. The biomechanical scaling factors for each region include the acceleration scaling factor and time scaling factor for the cranial region; the axial force scaling factor and bending moment scaling factor for the cervical spine region; the overall stiffness scaling factor for the thoracic region; the overall compressive force scaling factor for the abdominal region; the interface soft tissue stiffness scaling factor for the pelvic region; and the knee joint moment scaling factor for the lower limb region. The acceleration scaling factor and time scaling factor are obtained as follows: in, This represents the acceleration scaling factor for the cranial region. The time scaling factor representing the brain region. This represents the scaling factor for the shear modulus of soft tissue in the brain. This represents the scaling factor for the density of brain tissue. Indicates the geometric scaling factor of the brain; The axial force scaling factor and the bending moment scaling factor are obtained as follows: in, This represents the scaling factor for axial force in the cervical spine region. This represents the scaling factor for the bending moment in the cervical spine region. This represents the scaling factor for the elastic modulus of the cervical vertebrae. The geometric scaling factor representing the contact area between the cervical cortex and bone. This represents the scaling factor for the stiffness of the cervical ligaments. Geometric scaling factor representing the thickness of the cervical intervertebral disc; The overall stiffness scaling factor is obtained as follows: in, This represents the overall stiffness scaling factor for the thoracic cavity region. This represents the scaling factor for the elastic modulus of the ribs in the thoracic cavity. This represents the scaling factor for the thickness of the ribs in the thoracic cavity. Indicates the geometric scaling factor of the thoracic cavity; The overall compressive force scaling factor is obtained as follows: in, This represents the overall compressive force scaling factor for the abdominal region. This represents the first weighting coefficient related to BMI. This represents the second weighting coefficient related to BMI. The compressibility factor of a viscoelastic solid in the abdomen. This represents the scaling factor for the elastic modulus of abdominal organs. The geometric scaling factor representing the contact area of ​​the organ. The compressibility factor represents the compressibility factor of the incompressible fluid in the abdomen. This represents the scaling factor for the dynamic viscosity of the peritoneal fluid. This represents the scaling factor for the surface tension of the peritoneal fluid. The geometric scaling factor representing the inter-organ space; The interface soft tissue stiffness scaling factor is obtained in the following way: in, This represents the scaling factor for the interface soft tissue stiffness in the pelvic region. This represents the scaling factor for the shear modulus of the soft tissues surrounding the pelvis. This represents the ratio of the soft tissue thickness of the personalized geometric model to that of the baseline model, indicating the scaling factor for the soft tissue thickness around the pelvis. This represents the ratio of the interosseous distance of the pelvis in the personalized geometric model to that in the baseline model, indicating the pelvic geometry scaling factor. The knee joint torque scaling factor is obtained in the following way: in, This represents the knee joint torque scaling factor for the lower limb region. This represents the scaling factor for knee ligament stiffness. This represents the geometric scaling factor for the lower limbs.

7. A computing device, characterized in that, Including memory and processor; The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program for performing the method as described in any one of claims 1 to 5.

Images