A three-dimensional finite element model construction method for head chopping case research

Abstract

The present application relates to a kind of head hacking case research and judgment-oriented three-dimensional finite element model construction method, comprising: constructing personalized head geometric model, and with 3D printing technology to produce head physical model;Using head physical model and infrared light point motion capture system, carry out simulation hacking experiment and collect data, calculate to obtain wound morphology index;According to personalized head geometric model, establish head finite element model, computer simulation is carried out to hacking head wound formation;The simulation result is compared with wound morphology index, and the effectiveness of head finite element model is verified.The present application carries out simulation hacking experiment and finite element simulation experiment to hacking head, and the two are organically unified;Simulation hacking experiment effectively calibrates and verifies the simulation experiment result of finite element model, and the verified finite element model can repeatedly carry out computer simulation experiment, and the various possibilities of hacking case are verified by forward deduction, to realize reverse research and judgment.

Description

TECHNICAL FIELD

[0001] The application relates to a three-dimensional finite element model construction method for a head chopping case, and belongs to the technical fields of biomechanical modeling and computer simulation and computer visualization. BACKGROUND

[0002] The harm of a criminal using a knife to chop a victim is a common criminal behavior in civil disputes or terrorist attacks. After the case occurs, forensic investigators and other criminal investigators usually inspect the victim's chopping (wounds), combine the collected physical evidence on the scene, and comprehensively analyze and infer the case occurrence process, including the chopping behavior characteristics, the type of the knife, and the individual characteristics of the criminal suspect, to provide more complete evidence for case investigation, prosecution, and court trial.

[0003] In a chopping case, the morphological characteristics of the chopping, the identification of the chopping tool, the motion characteristics of the chopping behavior, and the individual characteristics of the harmdoer are four key elements of criminal investigation and court trial of the chopping case. Through scientific research, the quantitative relationship between the morphological characteristics of the chopping and the characteristics of the knife and the chopping behavior is established, which greatly improves the scientific level of criminal investigation and court trial of the chopping case.

[0004] The research team of the inventor has applied for multiple related patents and published related papers, as follows.

[0005] (1) The invention patent "Chopping injury case reproduction system and reproduction method" (application number CN202010278731.8, authorized publication number CN111523265B) was applied for on April 10, 2020. The system is composed of an input unit, a simulation analysis unit, and an output unit connected in turn. The input unit includes a case scene clue module and an artificial inference module. The simulation analysis unit includes a chopping knife digitization module, a chopping motion digitization module, a chopping human body multi-body system dynamics analysis module, and a human body chopped part finite element analysis module. The output unit includes a chopping event visualization module. The method uses the above system to reproduce and demonstrate in animation. The technical solution of the patent is based on the basic laws of human physiology, anatomy, and biomechanics. Through the calculation simulation technology and the computer visualization technology, the chopping injury case is reproduced, and the method has the advantages of being scientific, reliable, intuitive, realistic, and flexible.

[0006] (2) Invention patent application "Upper limb action space range calculation system, its construction method and use method" (application number CN202110112278.8, authorization announcement number CN112949031B) applied on January 27, 2021, in which the construction method conducts a simulation kinematics experiment of upper limb action with the subject as the action implementer, obtains kinematics data through motion capture of the upper limb action; establishes a whole body model; combines the human body model to analyze and process the kinematics data, calculates the dynamic change value of the three-dimensional space coordinates of the human body node marker point, the tool or non-tool instrument or the hand marker point in the upper limb action; calculates the upper limb action space range index; and establishes the upper limb action space range calculation system. The upper limb action space range calculation system established by the patent technical solution can calculate the injury person's upper limb action space range index, help criminal investigators to judge the compatibility of the injury person's action trajectory and the case scene space in the injury case, and provide a scientific basis for inferring the case evolution process.

[0007] (3) Invention patent application "Reverse judgment system for hacking formation, its construction method and judgment method" (application number CN202210220552.8, authorization announcement number CN114580241B) applied on March 8, 2022, in which the construction method is based on the biomechanics theory of human tissue damaged by hacking load, uses the traversal simulation experiment method of the finite element method to establish a reverse judgment scale of hacking formation factors, and further constructs a reverse judgment system; the reverse judgment system takes hacking characteristics and hacking angle, tool type and sharpness as input, and hacking speed and hacking force of hacking action as output, which is clear in meaning, easy to learn, and low in implementation difficulty. Using the patent technical solution, it is easy to combine with on-site investigation evidence to judge multiple possible situations, and with the aid of computer software, a cycle of judgment can be completed in a very short time, and multiple different factor combination judgments can be quickly made, which can greatly save time for case investigation or court evidence presentation.

[0008] (4) Published paper: Hao Weiya, Shan Songjie, Shi Yi, et al. Influence of sharpness of knife on forearm injury in hacking cases. Medical Biomechanics, 2020, 35(05):546-552. In which, by establishing a finite element model of knife and human forearm, the influence of sharpness of knife on forearm injury in knife hacking cases is quantitatively explored. The paper establishes a method for quantitatively evaluating the formation process of knife hacking wounds on the upper limbs. The research results deepen the understanding of the biomechanical mechanism of knife hacking wound formation, and provide a reference for forensic identification and court trial of hacking cases.

[0009] Based on the above technical means, the research team of the inventors has been able to determine the relationship between the action behavior characteristics of the person who commits the hacking (including hacking speed, hacking force, and action range), the characteristics of the person who commits the hacking (height, arm length, etc.), the characteristics of the hacking tool (chopper or kitchen knife), and the morphological characteristics of the wound of the victim (depth, circumference, etc.) in hacking cases through human motion capture technology, anthropometric data, human image data, and human three-dimensional finite element analysis methods.

[0010] However, the above research results are all described by taking hacking of the arm of the victim as an example, and in knife murder cases, the head is a common injury site, and the existing technical materials lack specific technical solutions for constructing and verifying the head finite element model, especially the above research results do not point out the technical solution for experimentally verifying the human finite element model constructed by medical imaging, so it is difficult to experimentally verify the head finite element model.

[0011] When a head hacking case occurs, criminal investigation and court trial need to reversely deduce the characteristics of the case, such as the hacking tool, hacking speed and hacking force level, and the characteristics of the suspect (such as gender, age, and physical characteristics), according to the wound (i.e., hacking wound) of the head of the victim and the scene evidence, to improve the correlation level of the case evidence and the scientific level of the case investigation and court trial. However, there is still a lack of technical means directly for hacking research and judgment of the head. SUMMARY

[0012] The purpose of the present application is to solve the above-mentioned problems of the prior art, based on the latest research results of the research group of the inventors, the present application proposes a three-dimensional finite element model construction method for hacking case research and judgment of the head, which can construct a head finite element model for hacking case research and judgment of the head based on the anatomical characteristics of the head.

[0013] The technical solution of the present application to solve the technical problems is as follows:

[0014] A three-dimensional finite element model construction method for hacking case research and judgment of the head, characterized by comprising the following steps:

[0015] Firstly, the scene of the head hacking case is investigated, and the data of the head of the victim is measured to construct a personalized head geometric model, and a head physical model is made by 3D printing technology.

[0016] Secondly, the head physical model and an infrared light point motion capture system are used to perform a simulation hacking experiment and collect data; a coordinate conversion relationship between the coordinate system of the infrared light point motion capture system and the coordinate system of the CAD system is obtained according to a preset fusion algorithm; the coordinate conversion relationship is used to process and calculate the collected data to obtain simulated wound morphological indexes; the simulated wound morphological indexes include wound circumference and wound depth.

[0017] Third step, according to the personalized head geometric model, the head finite element model is established and assembled with the cutter finite element model, and the computer simulation of the head wound formation is calculated; the simulation results include the wound circumference and the wound depth.

[0018] Fourth step, the simulation results obtained in the third step are compared with the simulated wound morphological indexes obtained in the second step, to verify whether the head finite element model is effective, if not, the head finite element model is calibrated and then the computer simulation step in the third step is turned to, if yes, the head finite element model is taken as the final constructed head finite element model.

[0019] The construction method is based on the personalized geometric model of the victim, two different types of models are established: 3D printing physical model and finite element model, by comparing the results of physical experiment and finite element simulation experiment, the validity of the finite element model is verified, and a scientific and effective finite element model is provided for further case research and judgment. (Note: the further case research and judgment can use the technical solutions of the applicant's existing patents CN111523265B and CN114580241B.)

[0020] The further improved technical solutions of the application are as follows:

[0021] Preferably, in the first step, the measured data of the victim's head includes: the position of the wound on the victim's head and the morphological indexes of the wound, and the three-dimensional geometric data of the victim's head; wherein the morphological indexes of the wound include the length and depth of the wound, and the three-dimensional geometric data includes the length, height and width of the head.

[0022] Preferably, in the first step, the specific process of constructing the personalized head geometric model is as follows:

[0023] The THUMS head model is imported into the computer aided engineering grid processing software, the model is scaled by 1:1 according to the three-dimensional geometric data of the victim's head, and the quality of 1D, 2D and 3D grid is checked respectively, if there are grid units that do not meet the standard, preset processing is carried out, the specific process of preset processing is: first, the local area of the head is redivided by using the grid automatic division function, then the unqualified units are deleted and qualified grid is regenerated; then the personalized head geometric model of the victim is obtained.

[0024] The specific process of making the head physical model by 3D printing technology is as follows:

[0025] The brain cranium, facial cranium and teeth of the personalized head model of the victim are reversely reconstructed; the brain cranium includes frontal bone, parietal bone, occipital bone, temporal bone, sphenoid bone and ethmoid bone; the facial cranium includes maxilla, mandible, zygomatic bone, lacrimal bone, nasal bone, palatine bone, inferior turbinate bone and vomer bone; during the reverse reconstruction, hole filling and smoothing and optimization processing are performed by using a curved surface processing software, and three-dimensional surface reconstruction is performed; then, the reconstructed model is imported into a medical image processing software to be converted into a 3D printing special model, and then a 3D printing technology is used to manufacture a physical head model with a size of 1:1.

[0026] More preferably, the material for printing the parietal bone of the head is polyether ether ketone (PEEK), and the material for printing the rest of the head is polylactic acid (PLA).

[0027] The above preferred technical solutions can further optimize the specific technical features of the first step, which is conducive to achieving better technical effects of the overall technical solution.

[0028] Preferably, in the simulated chopping experiment, the subject chops the predetermined part of the physical head model according to the predetermined chopping requirements by holding the preset knife; a plurality of geometric marker points are set on the subject, the physical head model and the preset knife respectively, and then a plurality of reflective marker points are set on the geometric marker points respectively; during the process of the subject chopping the physical head model, an infrared light point motion capture system continuously collects the spatial coordinates of each marker point at a preset sampling frequency.

[0029] Preferably, in the second step, the specific process of the preset fusion algorithm is as follows:

[0030] Suppose there are N marker points, and the spatial coordinates of each marker point at M time points are collected within a period of time; suppose the i-th marker point at the j-th time point is P ij i=1, 2, …, N; j=1, 2, …, M; suppose the coordinates of P ij in the coordinate system of the infrared light point motion capture system are and the coordinates of P in the coordinate system of the CAD system are The relationship between the two coordinates is formula (I):

[0031]

[0032] wherein, represents the rotation relationship between the two coordinate systems; represents the translation relationship between the two coordinates; solving the T matrix and the b matrix can realize the conversion of the coordinates in the above two coordinate systems;

[0033] The cost function of the optimization calculation is defined as:

[0034]

[0035] This transforms the solution to matrix T and matrix b into the solution to the optimal solution of equation (II), i.e.:

[0036]

[0037] make Then equation (III) becomes

[0038]

[0039] The steepest descent method is used for iterative solution to obtain the optimal solution. The specific process is as follows:

[0040] Define function J(T) cb The gradient of ) is:

[0041]

[0042] Then the function J(T) cb The gradient unit vector of ) is:

[0043]

[0044] in For function J(T) cb The magnitude of the gradient, i.e.:

[0045]

[0046] make:

[0047]

[0048] Where α k To find the optimal step size;

[0049] Using equation (VIII), iterative calculations are performed according to the following steps to obtain the solution function J(T). cb The optimal solution:

[0050] (i) Choose any initial point The convergence accuracy ε>0 is given;

[0051] (ii) Calculation The gradient and magnitude of the gradient at a point determine the search direction.

[0052] (iii) Determine whether the convergence criterion is met. If the condition is met, stop the iteration and output the optimal solution. Otherwise, proceed to the next iteration;

[0053] (iv) Starting from the point, one-dimensional search is carried out along the search direction, and after the optimal step is determined, the next iteration point is obtained Let k=k+1, and turn to step (ii) to continue iteration.

[0054] According to the obtained optimal solution, the T matrix and the b matrix are obtained, and the coordinate conversion relationship between the infrared light point motion capture system coordinate system and the CAD system coordinate system is obtained.

[0055] Preferably, in the second step, the specific process of obtaining the simulated wound morphology index is as follows: first, the chopping speed is obtained, then the spatial position of the tool plane at the end of the chopping and the spatial position of the tool edge at the end of the chopping are obtained, and then the wound circumference and the wound depth are calculated and obtained.

[0056] The chopping speed is the linear speed of the tool tip contacting the physical model of the head in the simulation chopping experiment, which is calculated by the infrared light point motion capture system based on the filtered spatial coordinates.

[0057] The spatial position of the tool plane at the end of the chopping is obtained as follows:

[0058] The data collected by the infrared light point motion capture system is imported into the CAD system using a preset fusion algorithm, and the position of the tool when the chopping speed is 0 is taken as the spatial position of the tool at the end of the chopping.

[0059] Three marker points on the tool are selected: the tool tip K-ST, the tool ridge K-BOL, and the center of mass K-C; and the spatial coordinates of these points in the CAD system are set as The spatial equation of the tool plane is:

[0060]

[0061] The formula is simplified as:

[0062] A K x+B K y+C K z+D K =0(X)

[0063] Formula (X) is the plane equation of the tool plane in the CAD system.

[0064] The spatial position of the tool edge at the end of the chopping is obtained as follows:

[0065] Let the spatial coordinates of the rear point K-B1 and the front point K-B2 of the tool edge be

[0066] For K-B1, according to K-B1 in the tool plane, the distance between K-B1 and the tool ridge K-BOL is the tool width d k, the line connecting K-BOL and K-B1 and the line connecting K-BOL and K-ST are mutually orthogonal and perpendicular, and a system of equations is set up as follows:

[0067]

[0068] According to the known coordinates of K-BOL and K-ST, the spatial coordinates of K-B1 are obtained by solving the system of equations (XI); and the spatial coordinates of K-B2 are obtained in the same way.

[0069] The spatial position of the cutting edge at the end of the chopping is approximated as the line connecting K-B1 and K-B2, and the equation for the spatial position of the cutting edge at the end of the chopping is determined as follows:

[0070]

[0071] The wound perimeter is obtained as follows:

[0072] The wound perimeter is the length of the surface curve of the head wound caused by the cutting tool at the end of the chopping; after the spatial position of the cutting tool at the end of the chopping is determined, the length of the profile curve formed by the intersection of the head surface and the tool plane is measured in the CAD system, and the wound perimeter is obtained; during the measurement, N Qk points Q i are set along the profile curve in the CAD system, and the spatial coordinates of the points are The interval between two adjacent points Q i is 1±0.3mm, and the length W L of the profile curve is approximated as the sum of the lengths of the broken lines composed of the points Q i , and the specific formula is as follows:

[0073]

[0074] The wound depth is obtained as follows:

[0075] The spatial position of the cutting edge at the end of the chopping is taken as the deepest part of the head wound; the perpendicular distance between each point Q Qk and the line connecting K-B1 and K-B2 is calculated in turn using the N i points Q i of the profile curve, and the maximum value is taken as the wound depth.

[0076] The specific process of the calculation is as follows: first, the vector S KB is set as:

[0077] S KB = (x KB2 -x KB1 )i + (y KB2 -y KB1 )j + (z KB2 -zKB1 )k=m KB i+n KB j+p KB k(XIV)

[0078] where i, j and k are unit vectors in the x, y and z directions respectively;

[0079] vector S KB is the distance between K-B1 and K-B2, i.e.

[0080]

[0081] Let the vector S i between K-B1 and Q KQ be:

[0082] S KQ =(x Qi -x KB1 )i+(y Qi -y KB1 )j+(z Qi -z KB1 )k=m QK i+n QK j+p QK k(XVI)

[0083] According to the method of spatial analytic geometry, the perpendicular distance between Q i and the line between K-B1 and K-B2 is:

[0084]

[0085] The numerator on the right side of the above equation is the modulus of the cross product of two vectors, and the denominator is the result of equation (XV);

[0086]

[0087] The modulus of the above vector is calculated according to the following equation:

[0088]

[0089] The perpendicular distance between each Q i point and the line between K-B1 and K-B2 is calculated in turn by equations (XV), (XIX) and (XVII), and the maximum value is taken as the depth of the wound.

[0090] The above preferred technical solutions can further optimize the specific technical features of the second step, which is beneficial to the overall technical scheme to achieve better technical effects.

[0091] Preferably, the specific process of establishing the skull finite element model according to the individualized skull geometry model and assembling the skull finite element model with the cutter finite element model in the third step comprises:

[0092] S1, using computer aided engineering grid processing software to perform unit grid division on the individualized skull geometry model to establish an individualized skull finite element model; in the establishment process, a layered modeling function is used to divide the skull into a plurality of 0.1mm-thick layers, and outer compact bone, middle cancellous bone and inner compact bone are respectively constructed to match the actual anatomical structure, wherein the thickness of the outer compact bone is 2.4±0.1mm, the thickness of the middle cancellous bone is 3.2±0.1mm, and the thickness of the inner compact bone is 2.4±0.1mm.

[0093] S2, in the computer aided engineering grid processing software, hexahedral elements are used to perform encryption processing on the unit grid of the contact area between the cutter and the skull, and progressive tetrahedral elements are used to process the grid of the transition area between the contact area and the non-contact area, and the main steps of the above processing comprise: using a grid subdivision function in the contact area between the cutter and the skull to encrypt the grid of the contact area, and the minimum grid size after encryption is 0.09mm; using a grid refinement strategy in the transition area between the contact area and the non-contact area to divide the grid of the transition area into progressive tetrahedral elements; the skull finite element model after the above processing comprises five categories of tissues, i.e. outer compact bone, middle cancellous bone, inner compact bone, gray matter and white matter.

[0094] S3, using a node merging strategy to process the grid common nodes of the skull finite element model, so that the skull finite element model has mechanical behavior consistency and geometric continuity.

[0095] S4, if the actual cutter has been found in the skull hacking case, a cutter finite element model is established according to the actual cutter, and the main steps thereof comprise: measuring the geometric parameters of the actual cutter using a vernier caliper, inversely reconstructing the outer contour of the actual cutter using the blade geometric data obtained by pre-scanning the preset cutter used in the second step of simulating the hacking experiment, applying a CAD system to establish a cutter geometric model, then importing the computer aided engineering grid processing software to perform grid division and local encryption processing on the blade area;

[0096] If the actual cutter has not been found in the skull hacking case, the possible cutter types in the case are pre-set, and then corresponding cutter finite element models are selected from the existing cutter database.

[0097] S5, in the finite element software, based on the hacking angle of the cutter in the second step of simulating the hacking experiment and the contact position between the cutter and the skull, the cutter finite element model and the skull finite element model are assembled.

[0098] S6, in the finite element software, set the parameters of each constitutive equation of the head finite element model and the parameters of the constitutive equation of the knife finite element model, the parameters include density, elastic modulus or Young's modulus, Poisson's ratio, ultimate stress and ultimate strain; according to the cadaver experiment data in the previous literature, each constitutive equation of the head finite element model is respectively assigned with corresponding material properties; the constitutive equation of the knife finite element model is assigned as a steel isotropic material.

[0099] More preferably, in the third step, the specific process of computer simulation of the formation of the chopping head wound includes:

[0100] The computer simulation is performed using the solver of the finite element software, the boundary conditions are determined by the chopping position, chopping angle and chopping speed obtained by simulating the chopping experiment in the second step, and the bottom of the head finite element model is fixed; in the head finite element model, the failure mode of the bone tissue unit is set to ductile fracture failure, and the failure mode of the soft tissue unit is set to shear failure, and during the computer simulation process, when the bone tissue unit and the soft tissue unit in the contact area of the knife and the skull are subjected to a load reaching the failure condition, the unit is set to fall off and removed from the calculation.

[0101] The above preferred technical solutions can further optimize the specific technical features of the third step, which is beneficial to the overall technical scheme to achieve better technical effects.

[0102] Preferably, the specific process of the fourth step is:

[0103] The wound circumference of the simulation result obtained in the third step is compared with the wound circumference of the wound morphological index obtained in the second step, and the wound depth of the simulation result obtained in the third step is compared with the wound depth of the wound morphological index obtained in the second step; if the difference is less than 15%, the head finite element model is determined to be effective; otherwise, the head finite element model is calibrated and then turned to the computer simulation step of the third step to obtain the simulation result again and continue the comparison until the head finite element model is determined to be effective, wherein the calibration measures include adjusting the unit mesh division and / or adjusting the boundary conditions; finally, the effective head finite element model is used as the final constructed head finite element model.

[0104] The above preferred technical solutions can further optimize the specific technical features of the fourth step, which is beneficial to the overall technical scheme to achieve better technical effects.

[0105] Compared with the prior art, the present application carries out a simulated chopping experiment and a finite element simulation experiment on the chopping head, and organically unifies the two. The simulated chopping physical experiment is limited, which effectively calibrates and verifies the simulation experiment results of the finite element model, and the verified finite element model can repeatedly carry out computer simulation experiments, verify various possibilities of the chopping case, and realize reverse research and judgment. BRIEF DESCRIPTION OF DRAWINGS

[0106] Figure 1 Flowchart for the implementation of the present application.

[0107] Figure 2 Schematic diagram of a 3D-printed human skull model with a 1:1 ratio in Example 1 of the present application, wherein (a) is a CAD skull model, and (b) is a 3D-printed skull model.

[0108] Figure 3 Schematic diagram of a 3D-printed skull model for chopping experiment testing in Example 2 of the present application.

[0109] Figure 4 Schematic diagram of the experimental kitchen knife and its blade geometric feature microstructure in Example 2 of the present application, wherein (a) is a schematic diagram of the appearance of the experimental kitchen knife, and (b) is an electron microscope measurement picture of the experimental kitchen knife. Note: the experimental kitchen knife in the figure contains acceleration and force sensors (see the applicant's utility model patent: "a motion and mechanical measurement tool for simulating chopping and stabbing", application number 202120686829.7, authorized publication number CN214372664U).

[0110] Figure 5 Schematic diagram of the chopping experiment process and data collection in Example 2 of the present application, wherein (a) is a schematic diagram of the chopping action of the 3D-printed skull, and (b) is a schematic diagram of the stick diagram generated by QTM software analysis.

[0111] Figure 6 Schematic diagram of the CAD model of the knife chopping the skull in Example 2 of the present application, wherein (a) is the initial contact time of chopping, and (b) is the time when the chopping knife stops moving and forms a chopping wound.

[0112] Figure 7 Position relationship diagram of the key points of the knife in Example 2 of the present application. Note: d k represents the width of the knife, d k = 9.0 cm.

[0113] Figure 8 Schematic diagram of the calculation of the length and depth of the chopping wound in Example 2 of the present application, wherein (a) is the deepest moment of the knife chopping, and (b) is the calculation of the length and depth of the wound circumference.

[0114] Figure 9 Schematic diagram of the actual measurement and calculation of the length and depth of the wound circumference of the 3D-printed skull in Example 2 of the present application.

[0115] Figure 10Fig. 3 is a schematic diagram of the finite element model of the head in Example 3 of the present application, wherein (a) is a schematic diagram of the layered structure of the finite element model of the head and the encryption region; (b) is a schematic diagram of the gray matter model of the head; and (c) is a schematic diagram of the white matter model of the head.

[0116] Figure 11 Fig. 4 is a schematic diagram of the finite element model of the kitchen knife in Example 3 of the present application.

[0117] Figure 12 Fig. 5 is a schematic diagram of the assembly of the finite element models of the kitchen knife and the head in Example 3 of the present application.

[0118] Figure 13 Fig. 6 is a schematic diagram of the computer simulation of the chopping of the head based on the finite element model in Example 3 of the present application, wherein (a) is a schematic diagram of the knife forming a wound; and (b) is a schematic diagram of the von Mises stress distribution of the skull. DETAILED DESCRIPTION

[0119] The flowchart of the embodiment of the present application is shown in Fig. 7. Figure 1 In the embodiment, the three-dimensional finite element model construction method for the head chopping case in the present application contains all the technical features (including the features of the preferred embodiments) described in the summary of the invention, and the model obtained by using the construction method is not repeated here.

[0120] It should be noted that the computer-aided engineering grid processing software, surface processing software, medical image processing software, finite element software and other software programs used in the present application are mature existing technology products, and those skilled in the art can easily find existing software programs with corresponding functions according to the functions of each software program described in the present application, and there are more than one product. Those skilled in the art can choose appropriate products, and the present application does not need to specify these software programs. This is a technical means that those skilled in the art can easily understand and easily implement, and there is no situation that the content described in the present application cannot constitute a clear and complete technical solution.

[0121] The present application will be further described in detail below with reference to the accompanying drawings and in conjunction with the embodiments. However, the present application is not limited to the examples given.

[0122] Example 1

[0123] In this embodiment, a physical model of the head is made by using 3D printing technology.

[0124] The specific content of this embodiment is as follows:

[0125] (1) On-site investigation

[0126] After the chopping case occurs, the scene is investigated, the position and morphological indicators of the wound on the victim's head are measured, including the length and depth of the wound. These indicators can be used for subsequent research and judgment after the finite element model is completed (see the applicant's existing patents CN111523265B and CN114580241B for details, which will not be repeated here).

[0127] The main morphological indicators of the victim's individualized head are measured using a ruler or vernier caliper, including the three-dimensional geometric data of the head (including length L, height H, width W, etc.).

[0128] (2) Construction of individualized head geometric model

[0129] According to the geometric data of the victim's head obtained by the above measurement, the publicly disclosed Total human model for safety (THUMS, Toyota Motor Corporation, Japan) is personalized to obtain the victim's individualized head geometric model. The specific process is as follows: import the THUMS head model into computer aided engineering grid processing software (for example: Gmsh, v4.11, University of Liège, Belgium) (Note: import directly or after preprocessing, specific according to the software used to determine), according to the geometric size of the victim's head, the model is scaled 1:1 in proportion. In this way, the geometric differences between the THUMS model and the victim's head can be minimized. www.toyota.co.jp / thums /

[0130] After scaling, the quality of 1D, 2D and 3D grids is checked respectively, and the unqualified grid elements are processed as follows: 1) the local area of the head is re-divided by using the grid automatic division function; 2) delete unqualified units and regenerate qualified grids.

[0131] After the scaling and quality check of the head model are completed, the victim's individualized head geometric model is obtained, and the model file is saved.

[0132] (3) 3D printing technology to make physical model

[0133] (3.1) Data model for 3D printing

[0134] ​A physical model of the skull was 3D-printed using a personalized geometric model of the victim's head. The model had similar geometric features to the victim's head, and the brain skull (frontal, parietal, occipital, temporal, sphenoid, and ethmoid) and face skull (maxilla, mandible, zygomatic, lacrimal, nasal, palatine, inferior turbinate, and vomer) and teeth were reverse reconstructed. The holes were filled and smoothed, and the model was optimized using a surface processing software (e.g., MeshLab, v2022.02, ISTI-CNR, Pisa, Italy), and a three-dimensional surface reconstruction was performed. The three-dimensional model was then imported into a medical image processing software (e.g., 3DSlicer, v5.4.0, Boston, MA, USA) and converted into a model for printing.

[0135] (3.2) Material selection

[0136] Two polymers were used in this example as materials for printing the skull, which had similar mechanical properties to human skull bones: 1) polyether ether ketone (PEEK) material was used to print the parietal bone of the skull (see Fig. b of Figure 2 , brown area in the figure); PEEK is a medical-grade material with similar mechanical properties to human skull bones; 2) polylactic acid (PLA) was used for the other parts of the skull and the support area (see Fig. b of Figure 2 , white area in the figure). These polymers were provided in the form of 3D-printing filaments, each with a diameter of 1.75 ± 0.05 mm. Table 1 presents the material properties of the printing filaments used.

[0137] Table 1, Material properties of the 3D-printed model and the real skull model

[0138]

[0139] (3.3) 3D skull printing process

[0140] The sample was separated from the parietal bone and other parts of the skull by the coronal suture, lambda suture and squamous suture based on the comprehensive consideration of experimental purposes, cost-effectiveness and experimental feasibility. The parietal bone and other parts of the skull were 3D printed with PEEK and PLA materials in a 1:1 size, respectively, and the printing method was fused deposition modeling (FDM) technology layer by layer printing. Considering that the 3D printed parietal bone and other parts of the skull need to bear the chopping action, the "mortise and tenon structure" was used to enhance the stability and strength between the structures, that is, the column (tenon head) was added on the left and right parietal bones, and the groove (mortise) was added on the underlying temporal bone. By inserting the tenon head into the mortise, the two parts were tightly connected. Table 2 shows the printing parameters used for 3D printing of the skull in this example.

[0141] Table 2, 3D printing parameter settings

[0142]

[0143] In addition, to enhance the adhesion between the printing bed and the molten layer and reduce the risk of warping, a special coating was applied to the printing bed. To ensure consistency of the final results, all samples were made following the same printing direction. After printing, the 3D printed skull was subjected to drying, ultraviolet curing, polishing and coloring engraving treatment, and the parietal bone and other parts of the skull were colored with yellow-brown and white, respectively.

[0144] The model of this example is shown in Figure 2 The model is from a 175 cm tall, 77 kg weight, 50% percentile body size adult male, and the model skull length x width x height: 152 mm x 200 mm x 256 mm. Considering that the skull needs to be fixed on the clamp in the experiment, a cuboid column (length x width x height: 50 mm x 75 mm x 110 mm) is added at the foramen magnum of the skull.

[0145] Example 2

[0146] This example uses the 3D printed skull model obtained in Example 1 to collect data through simulation of chopping experiments, processes the data using fusion algorithm, and calculates the wound morphology index.

[0147] The specific content of this example is as follows:

[0148] (1) 3D printed skull physical model chopping experiment

[0149] This experiment aims to provide verification data for the three-dimensional finite element model.

[0150] (1.1) Selection of the Slasher

[0151] According to the monitoring data published in the Fifth National Physical Fitness Monitoring Bulletin of China, the average height of young Chinese men (aged 20-29) is 172.4 cm and their weight is 71.6 kg. Based on this, the inventor's research team recruited a healthy, right-handed young man who was 173.0 cm tall and weighed 68.5 kg as a subject (i.e., the slasher).

[0152] (1.2) Instruments and Equipment

[0153] The impact test was conducted in a laboratory environment (length × width × height: 9.52m × 5.08m × 3.12m) equipped with an 8-lens 3D infrared light spot motion capture system (Oqus700, Qualisys TrackManager, Gothenburg, Sweden).

[0154] The 3D-printed skull model was fixed to a vise clamp (MYTEC, China), which was then secured to an adjustable landing gear (TN, China) with bolts. The height of the 3D-printed skull model could be adjusted by changing the height of the landing gear. To simulate the average height of an adult male as closely as possible, the height of the 3D-printed skull model was set to 1.70m. Two high-speed cameras (CASIO EX-F1, Japan) were positioned to the left and right of the subject to record images. Figure 3 ).

[0155] In this embodiment, a kitchen knife equivalent to a household kitchen knife was used as the test knife. The length, width, and weight of the knife were 35.8 cm, 9.0 cm, and 1.16 kg, respectively. Figure 4 (a) Figure). Since the sharpness of the blade has a direct impact on the wound morphology and severity of injury in slashing cases (see the paper mentioned in the background technology: Hao Weiya, Yamamatsu Junjie, Shi Yi, et al. The influence of the sharpness of the knife on forearm injury in slashing cases. Medical Biomechanics, 2020, 35(05):546-552). It is very important to accurately collect the geometry and sharpness of the blade, especially in finite element analysis where a fine model of the knife geometry is required. For this purpose, in this embodiment, a scanning electron microscope (SEM, JSM-IT200, JEOL, Japan) was used to measure the blade of the kitchen knife. The blade thickness and angle of the kitchen knife were 6.4 μm and 22.8°, respectively. Figure 4 (Figure (b)).

[0156] (1.3) Chopping test process and data collection

[0157] Before the experiment, participants were informed of the test content. They were asked to change into tight-fitting clothing and warm up. Reflective markers were then placed on the participants, the bony landmarks of the 3D-printed skull, and the geometric landmarks of the experimental kitchen knife (Table 3). Participants were instructed on the chopping motion and were allowed to adjust their standing position and orientation before striking the 3D-printed skull, depending on their individual circumstances. After understanding the chopping requirements, participants used the experimental kitchen knife to strike the right parietal bone of the 3D-printed skull with maximum force. Figure 5 (Figure (a)). During the chopping process, the chopping motion data was collected using an infrared light spot motion capture system (Qualisys Track Manager, QTM) at a sampling frequency of 250Hz. The spatial coordinates of the collected reflective markers were processed using the QTM system to obtain the spatial position data of the chopping tool speed, chopping spatial orientation, and wound circumference. Figure 5 (Figure (b)).

[0158] Table 3. List of Geometric Marker Points and Locations for Subjects, 3D Printed Skull Model, and Kitchen Knife

[0159]

[0160] (2) Motion capture data and CAD data fusion algorithm

[0161] The coordinates of the blade and reflective markers on the head during the strike can be obtained using the QTM motion capture system software, filtered using a 4th-order Butterworth low-pass filter with a cutoff frequency of 25Hz. To convert these coordinates to the 3D coordinates of the CAD software system, an algorithm for fusing the spatial coordinates from both systems needs to be developed.

[0162] (2.1) Calculation of coordinate transformation between motion capture data and CAD system

[0163] Coordinate transformations between different coordinate systems within the same space can be achieved by applying rotation and translation between coordinate systems. Let... Let A be the coordinates of point A in the QTM coordinate system, and Let A be the coordinates of point A within a CAD software system (e.g., FreeCAD, v0.21.2). According to spatial analytic geometry, the coordinates of point A in the two coordinate systems have the following relationship:

[0164]

[0165] in This represents the rotational relationship between two coordinate systems, while This represents a translation between two coordinate systems. According to spatial analytic geometry, the T matrix is ​​a 3rd-order symmetric invertible matrix. Clearly, if we obtain the T matrix and the b matrix, we can realize the transformation relationship between coordinates in two different coordinate systems.

[0166] (2.2) Solution algorithm of rotation matrix and translation matrix based on steepest descent method

[0167] In order to improve the conversion accuracy and stability between two coordinates, the embodiment is based on the least square method, and N points at different positions on the 3D printing head skull physical model (in order to improve the accuracy, when the N points are selected, points at different positions of the head skull can be selected) are selected, and the coordinates of the N points at M time points within a period of time t are obtained.

[0168] Since the sampling frequency of the motion capture system in the embodiment is 250 Hz, if the time t is selected as 1 s, there are 250 time points of three-dimensional coordinates for each marker point, and M=250.

[0169] Let be the coordinates of the P ij point (the i-th point at the j-th time point) in the QTM coordinate system, and be the coordinates of the point in the CAD software system. Then i=1, 2, …, N; j=1, 2, …, M.

[0170] According to formula (I'), the relationship between the above two coordinates is:

[0171]

[0172] wherein, represents the rotation relationship between the two coordinate systems; represents the translation relationship between the two coordinates. By solving the T matrix and the b matrix, the conversion relationship between the coordinates in the two different coordinate systems can be realized.

[0173] At this time, the cost function of the optimization calculation can be defined as

[0174]

[0175] In the above formula, the square root under the square root is the root mean square error between the calculation results and the actual results of all points at different times through coordinate conversion.

[0176] In this way, solving the T and b matrices becomes an optimization calculation problem of solving J(T, b). The values of the two matrices can be obtained by the most common optimization calculation method. Since the matrix T is a symmetric matrix, the T matrix has 6 independent variables (T 11 , T 12 , T 13 , T 22 , T 23 , T 33 ). In addition, the b matrix has 9 independent variables. In this way, the conversion between the motion capture space coordinate data and the CAD space coordinate data becomes the optimal solution of equation (II), that is,

[0177]

[0178] make Then formula (III) becomes

[0179]

[0180] The steepest descent method (also known as the optimal gradient method) is used for iterative solution to obtain the optimal solution. The algorithm is as follows:

[0181] Define function J(T) cb The gradient of ) is:

[0182]

[0183] Then the function J(T) cb The gradient unit vector of ) is:

[0184]

[0185] in For function J(T) cb The magnitude of the gradient, i.e.

[0186]

[0187] make:

[0188]

[0189] The above equation is the algorithm for iterative solution using the steepest descent method, where α k The optimal step size is determined by the algorithm described above. After constructing the algorithm, the function J(T) can be solved. cb The optimal solution T) cb The specific iterative calculation steps include:

[0190] (i) Choose any initial point The convergence accuracy ε>0 is given;

[0191] (ii) Calculation The gradient and magnitude of the gradient at a point determine the search direction.

[0192] (iii) Determine whether the convergence criterion is met. If the condition is met, stop the iteration and output the optimal solution. Otherwise, proceed to the next iteration;

[0193] (iv) Starting from a point, perform a one-dimensional search along the search direction. After determining the optimal step size, obtain the next iteration point. Let k = k + 1, go to step (ii) to continue iteration.

[0194] By the above steps, the optimal solutions of formula (III) and (IV) are obtained, so as to obtain the rotation matrix T and the translation matrix b between the QTM coordinate system and the CAD system coordinate system, and realize the coordinate conversion calculation between the motion capture data and the CAD system.

[0195] (3) Measurement and calculation of wound morphology index

[0196] (3.1) Chop speed

[0197] The chop speed is defined as the linear speed of the knife tip contacting the 3D printed skull model in the chop experiment, which is obtained by calculating the filtered spatial coordinates by the QTM system (as shown in the moment of the a figure of Figure 6

[0198] (3.2) Spatial position of the knife plane at the end of the chop

[0199] The dynamic change of the knife tip speed during the whole chop process can also be obtained in the QTM system. Obviously, when the knife speed becomes 0, the knife stops moving, and at this time, the chop wound is formed. In order to measure the wound morphology index, it is necessary to first determine the spatial position relationship between the knife and the skull at the end of the chop (as shown in the moment of the b figure of Figure 6

[0200] Using the motion capture data and CAD system coordinate conversion calculation method described above, the motion capture data is imported into the CAD system to determine the spatial position of the knife at the maximum chop depth moment (the position when the chop speed is 0).

[0201] Select the spatial coordinates of three points (K-ST, K-BOL and K-C) on the knife in the CAD system According to spatial analytic geometry, the spatial equation of the knife plane is determined as determinant (IX):

[0202]

[0203] The above equation can be simplified as:

[0204] A K x+B K y+C K z+D K = 0 (X)

[0205] In this way, the plane equation of the knife plane in the CAD system is established.

[0206] (3.3) Spatial position of the knife edge at the end of the chop

[0207] Let​​Figure 7 The spatial coordinates of the two points K-B1 and K-B2 on the blade of the middle cutter are respectively and To obtain the spatial coordinates of the point K-B1, the following three conditions can be used to set up an equation group: according to the point K-B1 in the cutter plane, the distance between the point K-B1 and the point K-BOL is d k (Note: d k represents the width of the cutter, d k = 9.0 cm), the line connecting the point K-BOL and K-B1 is perpendicular to the line connecting the point K-BOL and K-ST Figure 7 ), that is, the following equation group exists:

[0208]

[0209] Since the coordinates of the points K-BOL and K-ST are known, the above equation group has a unique solution, and after solving, the spatial coordinates of the point K-B1 can be obtained; in the same way, the spatial coordinates of the point K-B2 can also be calculated.

[0210] The spatial position of the blade at the end of the chopping can be approximately regarded as the line connecting the two points K-B1 and K-B2. According to spatial analytic geometry, the equation for determining the spatial position of the blade at the end of the chopping is:

[0211]

[0212] (3.4) Wound perimeter calculation

[0213] The wound perimeter refers to the length of the surface curve of the head wound caused by the cutter at the end of the chopping (as shown in the b figure of Figure 8 ). After determining the spatial position of the cutter plane, the length of the contour curve formed by the intersection of the head surface in the head model and the cutter plane can be measured in the CAD software system. Simply, the point spatial coordinate data with an interval of about 1 mm is generated by manually clicking the mouse in the CAD software system. It is assumed that there are N Qk points Q i ( Figure 8 ), and the spatial coordinates are In this way, the length W L of the wound contour curve can be approximately calculated according to the sum of the lengths of the polygonal lines composed of Q i , which is specifically as follows:

[0214]

[0215] (3.5) Wound depth calculation

[0216] The spatial position reached by the blade at the end of the chopping is basically the deepest part of the wound Figure 8). The distance between the two points K-B1 and K-B2 on the cutting edge at the end of the stroke is calculated by the following equation (XII). i The perpendicular distance between the two points K-B1 and K-B2 on the cutting edge at the end of the stroke is calculated by the following equation (XII).

[0217] To calculate the perpendicular distance between the two points K-B1 and K-B2 on the cutting edge at the end of the stroke, first set the vector S i KB

[0218] S KB = (x KB2 -x KB1 )i + (y KB2 -y KB1 )j + (z KB2 -z KB1 )k = m KB i + n KB j + p KB k (XIV)

[0219] where i, j and k are unit vectors in the x, y and z directions respectively.

[0220] The modulus of the vector S KB is the distance between the two points K-B1 and K-B2 on the cutting edge at the end of the stroke, i.e.

[0221]

[0222] Let the vector S i between K-B1 and Q KQ be:

[0223] S KQ = (x Qi -x KB1 )i + (y Qi -y KB1 )j + (z Qi -z KB1 )k = m QK i + n QK j + p QK k (XVI)

[0224] According to the method of spatial analytic geometry, the perpendicular distance between the two points K-B1 and K-B2 on the cutting edge at the end of the stroke is: i

[0225]

[0226] The numerator on the right side of the above equation is the modulus of the cross product of two vectors, and the denominator is the result of equation (XV).

[0227] ​​​

[0228] The modulus of the above vector is calculated by the following formula:

[0229]

[0230] Thus, Q can be calculated by formula (XV), (XIX) and (XVII) in turn. i The vertical distance from the point to the line connecting K-B1 and K-B2, and the maximum value is the depth of the wound d. This index can be used to verify the results of subsequent finite element simulation.

[0231] Example: Wound circumference and depth calculation

[0232] The research group conducted a chopping experiment on a 3D printed head model by a male youth. In the experiment, a three-dimensional motion capture system (Oqus700, Qualisys Track Manager, Gothenburg, Sweden) was used to collect the dynamic changes of the 3D spatial positions of the chopper, the knife and the infrared marker points on the head. The head physical model was fixed on a bench clamp (MYTEC, China), and the bench clamp was fixed on an adjustable landing gear through bolts. By adjusting the height of the landing gear, the height of the 3D head can be adjusted.

[0233] The collected data was processed by the above "(2) Motion capture data and CAD data fusion algorithm", and finally 54 point data of the wound curve were obtained through the CAD software system ( Figure 9 ), and the 3D spatial coordinates of each point were as follows:

[0234] Q1 = (289.29, 280.58, 1645.49),

[0235] Q2 = (289.61, 281.04, 1646.40),

[0236]

[0237] Q 54 = (322.25, 295.53, 1605.71).

[0238] 1. Wound circumference calculation

[0239] According to formula (XIII), the wound circumference is calculated,

[0240]

[0241] After substituting the coordinates of each point, we get:

[0242] W L ≈ 52.3 mm.

[0243] ​2. Wound depth calculation

[0244] The line connecting the two ends of the wound coincides with the straight line of the knife blade, so the wound depth can be calculated using the two points Q1 and Q2 and the vector S i (i = 2,..., 53) to find the maximum perpendicular distance from Q i to the line connecting Q1 and Q2.

[0245] S KB = Q 54 - Q1 = 32.96i + 14.95j - 39.78k

[0246] According to formulas (XV) to (XIX), the distance of Q i (i = 2,..., 53) to the vector S KB can be calculated, and the maximum value is the depth of the wound.

[0247] Specifically, Q 26 = (310.51, 284.26, 1629.87), then

[0248] S Q26 = Q 26 - Q1 = 21.23i + 3.68j - 15.62k

[0249] Then, according to formula (XVIII), d Q26 is calculated as the vector product of S KB and S

[0250] S KB x S Q26 = -86.95i - 330.97j - 196.06k

[0251] The modulus of the above two vectors is calculated respectively, and the result is

[0252]

[0253] Finally, substituting formula (XVII) can obtain the wound depth at Q 26 point:

[0254]

[0255] Example 3

[0256] Based on Example 1 and Example 2, this example establishes a finite element model for the head, performs computer simulation of the formation of the wound caused by chopping the head, and verifies the finite element model of the head.

[0257] The specific content of this example is as follows:

[0258] (1) Establishment of the finite element model of the head

[0259] (1.1) Individualized head finite element model structure reconstruction and element division

[0260] Based on the case victim head geometry model established in "(2) Individualized head geometry model construction" of Example 1, the structure is refined and discretized to construct an individualized head finite element model. The specifics are as follows:

[0261] Since the existing THUMS model is mainly applied to macro-scale damage research such as automobile and train collisions, the bone compact and spongy structure of the head is not considered in the model construction. Therefore, the model based on THUMS cannot accurately analyze and calculate the damage caused by sharp instrument injury to the local head. In this embodiment, the "sandwich" structure of "outer compact bone-spongy bone-inner compact bone" of the skull is finely modeled in the computer-aided engineering grid processing software, with an average thickness of 6.4 mm (of which, outer compact bone: 1.82 mm; inner compact bone: 1.75 mm; intermediate spongy bone: 2.83 mm) to enhance the accuracy of the model in local damage assessment. After completing the head geometry modeling, the computer-aided engineering grid processing software is used to further element grid division of the head three-dimensional geometry model to establish the head finite element model. In the establishment process, the skull is divided into several 0.1 mm thick levels using the layered modeling function, and the thickness of the compact bone and spongy bone is constructed to match the actual anatomical structure (see Table 4) according to the cadaver experiment data in the reference.

[0262] Table 4, tissue structure and finite element model

[0263]

[0264] In order to improve the calculation accuracy and reduce the number of elements in the whole model, in the discretization process of the head geometry model, the hexahedral elements are used for encryption processing in the area subjected to chopping (such as the frontal bone, left and right parietal bone and occipital bone area), with the minimum size reaching 0.09 mm, and the other parts use larger size elements, which effectively speeds up the requirement of the finite element model on computer hardware and also speeds up the solving speed.

[0265] When the head is injured by a knife, the skull and skin are the main load-bearing tissues, and the internal structure of the head (cerebrospinal fluid, white matter and gray matter, etc.) plays a smaller role. Therefore, these tissues use larger size elements in element division to reduce the number of elements and speed up the finite element model solving speed. The head finite element model after element division is shown in Figure 10 .

[0266] (1.2) Knife finite element model structure reconstruction and element division

[0267] The finite element model of the knife can refer to the existing patent technical solution of the inventor's research team (Hao Weiya, Xiao Xiaofei. Reverse judgment system for knife wound formation, construction method and judgment method [P]. Beijing: CN202210220552.8, 2024-10-01.). The general steps include: selecting an actual knife as a representative prototype, measuring the geometric parameters with a vernier caliper, and then applying CAD software to establish a geometric model of the knife after reverse reconstruction of the outer contour of the knife based on the geometric data of the knife edge obtained by scanning electron microscopy in Example 2. After completing the geometric modeling, the model is exported and saved in STP format. Import the knife model in STP format into computer-aided engineering mesh processing software for mesh division. In particular, the cutting edge area is locally encrypted to ensure the accuracy of the simulation. For example, Figure 11 The finite element model of a kitchen knife includes 53,304 elements and 10,656 nodes.

[0268] When there is no actual knife, the types of possible knives in the case can be pre-set, and then the corresponding knife finite element model is selected from the knife database (see patent: Hao Weiya, Xiao Xiaofei. Reverse judgment system for knife wound formation, construction method and judgment method [P]. Beijing: CN202210220552.8, 2024-10-01.) as the knife model of this example.

[0269] (1.3) Mesh processing of contact area and transition area

[0270] In the computer-aided engineering mesh processing software, after the skull is layered, in order to more accurately capture the wound changes of the contact area between the knife and the skull (in this example, the frontal bone is taken as an example), the hexahedral elements are used to encrypt the mesh (skin, cancellous bone and compact bone) of the contact area (see Figure 10 ). The operation steps are as follows: determine the contact position of the knife and the skull according to the experimental data obtained by the three-dimensional motion capture system in the chopping experiment, then select the contact area and use the mesh subdivision function to encrypt the mesh of the contact area. The minimum mesh size after encryption is 0.09mm. In order to optimize the calculation amount, the non-contact area is not encrypted. In the transition area between the contact area and the non-contact area, the mesh refinement strategy is used to divide it into a gradual tetrahedral element. Finally, the skull finite element model includes five types of tissues: outer compact bone, middle cancellous bone, inner compact bone, gray matter and white matter. The model has two types of elements: hexahedron and tetrahedron, with a total of about 4.04 million elements (see Table 4 in the above).

[0271] (1.4) Model assembly

[0272] Since the original skull is divided into three parts, and there is no common node between the parts, the node merging strategy is used to handle the grid common node to ensure the mechanical behavior consistency and geometric continuity of the model. After completing the meshing, save the model file, and also import the tool model file into the computer aided engineering grid processing software. Finally, in the finite element software (for example: MSC Marc (v2007, MSC Software Corporation, USA)), based on the chopping angle and contact position of "tool-skull" obtained by the three-dimensional motion capture system in the chopping experiment, the model is assembled after assembly, as shown in Figure 12 .

[0273] (1.5) Material properties of the head and tool model

[0274] In finite element simulation, selecting a suitable constitutive equation for the head tissue is one of the important factors affecting the calculation accuracy. In this embodiment, according to the cadaver experiment data in the previous literature, the material parameters of each part and each tissue of the human head are assigned.

[0275] Specifically, the parameters of each constitutive equation are set in the finite element software, including density, elastic modulus (Young's modulus), Poisson's ratio, ultimate stress and ultimate strain, etc.; then, according to the literature data, these constitutive equations are assigned to the material properties of each component of the tool and the head to ensure the physical consistency and accuracy of the simulation.

[0276] The tool is set as a steel isotropic material, and a linear elastic model in the finite element software is used for modeling. The scalp is also modeled using a linear elastic model. The bone parts of the head, such as the skull, facial bones and mandible, are modeled using an elastic-viscous-plastic material model. A viscoelastic material model is used to simulate the cerebrospinal fluid and the gray matter and white matter in the brain tissue (Table 5).

[0277] Table 5, Mechanical properties of each tissue in the tool and finite element head model

[0278]

[0279] Note: E, elastic modulus; σγ, yield stress; K, bulk modulus; G0, initial shear modulus; GI, infinite shear modulus; λ, attenuation constant.

[0280] (2) Computer simulation of the formation of the head wound by chopping

[0281] (2.1) Boundary conditions

[0282] After the construction of the head finite element model is completed, and the material mechanics properties and assignment of each tissue (compact bone, cancellous bone, cerebrospinal fluid and nerve tissue) in the model and the knife are determined, the solver of the finite element software can be used to perform the head chopping computer simulation experiment. The boundary conditions of the computer simulation experiment are determined by the parameters such as the chopping position, chopping angle and chopping speed obtained from the chopping 3D printed head simulation experiment of Example 2. The bottom (neck) of the head finite element model is fixed. These conditions constitute the boundary conditions of the finite element model.

[0283] (2.2) Computer simulation of wound formation

[0284] The research team of the inventors has published a paper (Hao WY, Yamamatsu JK, Shi Y, et al. Influence of sharpness of knife on forearm injury in chopping cases. Medical & Biological Engineering & Computing, 2020, 35(05): 546-552) on the chopping of the human forearm by a knife. This embodiment follows the research on the damage mode and theory of bone tissue and soft tissue under the action of the knife chopping. The failure mode of bone tissue is set to ductile fracture failure, and the failure mode of soft tissue is set to shear failure.

[0285] During the computer simulation, when the load on the bone tissue and soft tissue units in the cutting area (i.e., the contact area between the knife and the skull) reaches the failure condition, the unit is set to fall off by the software and removed from the calculation.

[0286] (3) Finite element model verification

[0287] Compare the wound circumference and wound depth formed by the chopping 3D printed head simulation experiment and the finite element model simulation experiment. If the difference is less than 15%, the finite element model is considered effective. If ≥ 15%, the finite element model is calibrated by adjusting the unit division, boundary conditions and other measures of the model, so that the difference between the finite element simulation results and the chopping 3D printed simulation experiment results is within 15%.

[0288] Example: This example combines a case to illustrate the implementation process of the technical solution of the present application in constructing a personalized finite element model and its application. The basic situation of the case is that a male resident in a residential area was killed by a knife in the living room of his home, and a kitchen knife was left at the scene. The perpetrator escaped, but the police locked in a suspect who was a 1.73m tall, 74kg, and physically strong man. The case investigation needs to analyze the crime process, especially the chopping process; confirm whether the characteristics of the murderer are consistent with the suspect.

[0289] According to the technical solution of the present application, the following steps are taken to construct the head finite element model of the case:

[0290] T1, After the case, the police lawfully carried out on-site investigation, to determine the identity of the victim, hit the knife and the identity of the suspect and its physical characteristics.

[0291] T2, the victim's head key geometric dimensions were measured, and the victim's head length x width x height was 152 mm x 200 mm x 256 mm.

[0292] T3, according to the victim's key geometric dimensions, the THUMS model was scaled, and the victim's personalized head geometric model was constructed in CAD.

[0293] T4, based on the victim's personalized head model in CAD software, 3D printing was carried out to make a 3D printed physical model of the head.

[0294] T5, find a strong man similar in height and weight to the suspect as a test volunteer, and simulate the chopping experiment on the 3D printed physical model. An infrared motion capture system was used to collect the dynamic changes of the three-dimensional space coordinate values of the infrared reflective marker points adhered to the human body, the knife and the physical model of the head.

[0295] T6, the motion capture data and CAD data fusion calculation was carried out to solve the conversion and translation of the three-dimensional space coordinates in the motion capture coordinate system and the three-dimensional space coordinate system in the CAD software system, and realize the mutual conversion of the two kinds of space coordinates.

[0296] T7, the chopping speed in the volunteer's simulated chopping 3D printed head physical model experiment was extracted by QTM system as 8.7 m / s; then the spatial position of the knife plane and the knife edge at the end of the chopping were determined in turn, and finally the wound circumference was calculated as 54.2 mm and the depth was 7.9 mm.

[0297] T8, the element quality in the scaled head model based on the THUMS model was checked in the computer aided engineering grid processing software, and the ill-conditioned elements caused by scaling were eliminated; the geometric model of the on-site knife was constructed, and the corresponding finite element model was established; the grid in the contact area of the knife and the head was refined; then the model assembly was carried out, and the knife model and the head model were assembled in the finite element software; finally, the parameters of the material properties of the head and the knife in the software were checked.

[0298] T9, using the data of the chopping site and the chopping speed from the chopping simulation experiment, the solver of the finite element software was used to calculate the wound shape and the stress distribution characteristics of the head caused by the chopping.

[0299] T10, the wound circumference and depth obtained by computer simulation were within ± 15% of the actual simulation experiment results (circumference: 46.1-62.3 mm; depth: 6.7-9.1 mm); the results showed that the constructed finite element model was effective.

[0300] T11, after verifying the validity of the finite element model, the technical solutions of the applicant's existing patents CN111523265B and CN114580241B can be used for further computer simulation experiments on the victim's hacking site and the suspect, to assist criminal investigation and court technical personnel in judging various possibilities of the case.

[0301] Based on the above embodiments and examples, the main key points of the technical solutions of the application are:

[0302] 1. To establish a finite element model for hacking assistance judgment, the application is based on the public THUMS finite element model, and the THUMS model is scaled according to the geometric characteristics of the victim's head to build a personalized geometric model of the victim.

[0303] 2. Based on the personalized geometric model of the victim, two different types of models are established: 3D printed physical model and finite element model. By comparing the results of physical experiments and finite element simulation experiments, the effectiveness of the finite element model is verified, and a scientific and effective finite element model is provided for further case judgment (the technical solutions of the applicant's existing patents CN111523265B and CN114580241B can be used for further case judgment).

[0304] 3. The first time to build a finite element model of the head under the condition of knife hacking. The head is different from other structures such as the forearm, and the head has skin, skull and intracranial tissue; the skull has a three-layer structure of compact bone-cancellous bone-compact bone, and the skull has nerve tissue structures such as gray matter and white matter.

[0305] 4. Motion capture data and CAD data fusion algorithm. In order to realize the mutual conversion of three-dimensional space of motion capture data and three-dimensional space of CAD data, through spatial analytic geometry and steepest descent method (also known as optimal gradient method) for iterative solution, the rotation and translation matrix between the two space coordinate systems is obtained.

[0306] 5. Wound morphology index measurement and calculation method. Based on spatial analytic geometry and curve length integral calculation method, ① the length of the line segment formed by the scattered points of the wound is summed up, and the wound circumference is calculated; ② find the maximum value of the perpendicular distance between the scattered points and the straight line of the blade as the hacking depth.

[0307] 6. Infrared light point pasting technical solution in hacking head simulation experiment, including the pasting position of infrared light points of human body, knife, 3D printed head and dummy.

[0308] The specific advantages of the application are:

[0309] 1. This invention is the first to conduct simulated slashing physics experiments and finite element simulation experiments on decapitation, and organically unifies the two. The simulated slashing physics experiment, as a finite experiment, effectively calibrates and verifies the results of the finite element model simulation experiment. The verified finite element model can then be repeatedly subjected to computer simulation experiments to forward extrapolate and verify various possibilities of slashing cases, thereby achieving reverse analysis.

[0310] 2. In the development and processing of the 3D printed skull physical model of this invention, the mechanical properties of the material used are similar to those of human bone tissue, and the printed geometry is processed according to the CAD geometric model. This realizes the homologous heterogeneous properties of the finite element model and the physical model, forming two different modal models, which provides extremely favorable conditions for physical model experiments.

[0311] 3. Based on spatial analytical geometry, this invention proposes a fusion and mutual conversion algorithm that integrates spatial coordinates from motion capture systems and CAD systems.

[0312] 4. Based on spatial analytical geometry and calculus, this invention proposes a method for collecting and calculating morphological data of the slash wound in a simulated head-slashing physical model experiment.

[0313] 5. A finite element model of the skull under knife-wielding conditions was constructed for the first time. The skull has a structure different from other structures such as the forearm. The skull has skin, skull bone and intracranial tissue; the skull bone has a "sandwich" structure of compact bone-cancellous bone-compact bone, and the skull bone contains gray matter and white matter structures.

[0314] 7. This is the first time that the finite element model has been used to simulate the damage and failure of human skin, bone, and nerve tissue under knife strikes, which is highly innovative.

[0315] 8. In constructing the finite element model, this invention refines the elements in the impact region and gradually increases the element size as the model moves away from the impact region. This element partitioning strategy ensures both computational accuracy and model continuity, while also improving computational speed.

[0316] 9. In the process of constructing the geometric model of the experimental knife, the geometric dimensions of the blade were observed and measured for the first time using an electron microscope, ensuring the authenticity of the constructed finite element model.

[0317] In addition to the embodiments described above, the present invention may have other implementations. All technical solutions formed by equivalent substitution or equivalent transformation fall within the protection scope claimed by the present invention.

Claims

1. A method for constructing a three-dimensional finite element model for the analysis of cases involving head injuries, characterized in that, Includes the following steps: The first step is to conduct an on-site investigation of the head injury case, measure and obtain the data of the victim's head, construct a personalized geometric model of the head based on this data, and then use 3D printing technology to create a physical model of the head. The second step involves conducting a simulated chopping experiment and collecting data using a physical model of the skull and an infrared light spot motion capture system. A pre-defined fusion algorithm is used to obtain the coordinate transformation relationship between the infrared light spot motion capture system coordinate system and the CAD system coordinate system. This coordinate transformation relationship is then used in conjunction with the collected data to process and calculate the simulated wound morphology indicators. These simulated wound morphology indicators include wound circumference and wound depth. The third step involves establishing a finite element model of the skull based on the personalized skull geometry model and assembling it with the finite element model of the blade to perform computer simulation of the formation of the skull wound from the chopping motion; the simulation results include the wound circumference and wound depth. The fourth step is to compare the simulation results obtained in the third step with the simulated wound morphology indicators obtained in the second step to verify whether the finite element model of the skull is effective. If not, the finite element model of the skull is calibrated and then the computer simulation step in the third step is performed. If it is effective, the finite element model of the skull is used as the final finite element model of the skull.

2. The method for constructing a three-dimensional finite element model for analyzing head injury cases according to claim 1, characterized in that, In the second step, during the simulated chopping experiment, the subject uses a pre-set knife to chop at a predetermined part of the physical head model according to the predetermined chopping requirements. Several geometric markers are set on the subject, the physical head model, and the pre-set knife in advance, and then reflective markers are set on each geometric marker. During the process of the subject chopping the physical head model, an infrared light spot motion capture system is used to continuously collect the spatial coordinates of each marker at a preset sampling frequency.

3. The method for constructing a three-dimensional finite element model for analyzing head injury cases according to claim 2, characterized in that, In the second step, the specific process of the preset fusion algorithm is as follows: Suppose there are N marker points, and the spatial coordinates of each marker point are collected at M time points over a period of time; let the coordinate of the i-th marker point at time j be P. ij Points; i = 1, 2, ..., N; j = 1, 2, ..., M; Let P ij The coordinates of the point in the infrared light point motion capture system coordinate system are: And the coordinates of this point in the CAD system coordinate system are The relationship between the two coordinates is given by equation (I): in, This represents the rotational relationship between two coordinate systems; This represents the translation relationship between two coordinate systems; by solving the T matrix and the b matrix, the coordinate transformation between the two coordinate systems can be achieved. Define the cost function for optimization computation: This transforms the solution to matrix T and matrix b into the solution to the optimal solution of equation (II), i.e.: (T * ,b * )=arg min T,b J(T,b) (III) make Then equation (III) becomes The steepest descent method is used for iterative solution to obtain the optimal solution. The specific process is as follows: Define function J(T) cb The gradient of ) is: Then the function J(T) cb The gradient unit vector of ) is: in For function J(T) cb The magnitude of the gradient, i.e.: make: Where α k To find the optimal step size; Using equation (VIII), iterative calculations are performed according to the following steps to obtain the solution function J(T). cb The optimal solution: (i) Choose any initial point The convergence accuracy ε>0 is given; (ii) Calculation The gradient and magnitude of the gradient at a point determine the search direction. (iii) Determine whether the convergence criterion is met. If the condition is met, stop the iteration and output the optimal solution. Otherwise, proceed to the next iteration; (iv) Starting from a point, perform a one-dimensional search along the search direction. After determining the optimal step size, obtain the next iteration point. Let k = k + 1, then proceed to step (ii) to continue the iteration; Based on the obtained optimal solution, the T matrix and b matrix are derived, and the coordinate transformation relationship between the infrared light spot motion capture system coordinate system and the CAD system coordinate system is obtained.

4. The method for constructing a three-dimensional finite element model for analyzing head injury cases according to claim 2, characterized in that, In the second step, the specific process of obtaining the simulated wound morphology indicators is as follows: first, obtain the cutting speed, then obtain the planar spatial position of the blade at the moment of the last cut, and the spatial position of the blade edge at the moment of the last cut, and then calculate the wound circumference and wound depth. The striking speed is the linear velocity of the blade tip at the instant the blade contacts the physical model of the head in the simulated striking experiment, which is calculated by the infrared light spot motion capture system based on the filtered spatial coordinates. The planar spatial position of the tool at the final instant of the strike is obtained through the following process: Using a preset fusion algorithm, the data collected by the infrared light spot motion capture system is imported into the CAD system. The position when the cutting speed is 0 is taken as the instantaneous spatial position of the maximum cutting depth, which is the planar spatial position of the cutting tool at the moment of the last cutting. Select three marker points on the cutting tool: tool tip K-ST, tool ridge K-BOL, and centroid KC; let the spatial coordinates of these points in the CAD system be... If i = 1, 2, 3, then the spatial equation of the tool plane is: The simplified expression is: A K x+B K y+C K z+D K =0 (X) Equation (X) is the plane equation of the tool plane in the CAD system; The spatial position of the blade at the final instant of the strike is obtained as follows: Let the spatial coordinates of the tool's back edge point K-B1 and front edge point K-B2 be respectively For K-B1, since K-B1 lies within the tool plane, the distance between K-B1 and the tool ridge K-BOL is the tool width d. k The lines connecting the ridge K-BOL and K-B1 and the line connecting the ridge K-BOL and the tip K-ST are orthogonal and perpendicular to each other. Establish a system of equations: Based on the known K-BOL coordinates of the blade spine and K-ST coordinates of the blade tip, the spatial coordinates of K-B1 are obtained by solving the system of equations (XI); similarly, the spatial coordinates of K-B2 are obtained. Approximating the spatial position of the blade at the final instant of the strike as the line connecting K-B1 and K-B2, the equation for the spatial position of the blade at the final instant of the strike is determined as follows: The wound circumference was obtained through the following process: The wound circumference is the length of the surface curve of the skull wound caused by the blade at the last instant of impact. After determining the spatial position of the blade plane at the last instant of impact, the length of the contour curve formed by the intersection of the skull surface and the blade plane is measured in the CAD system, thus obtaining the wound circumference. During measurement, N is set along the aforementioned contour curve in the CAD system. Qk Q points i And its spatial coordinates are Two adjacent points Q i The interval is 1±0.3mm, and the length W of the above contour curve is... L Press Q i The sum of the lengths of the resulting broken lines can be approximated using the following formula: The wound depth is obtained as follows: The spatial position of the blade at the moment of impact is taken as the deepest part of the skull wound; using the N of the aforementioned contour curve... Qk Q points i Calculate Q at each point in sequence. i The maximum value of the vertical distance between the lines connecting K-B1 and K-B2 is the wound depth. The specific calculation process is as follows: First, let vector S be... KB for: S KB =(x KB2 -x KB1 )i+(y KB2 -y KB1 )j+(z KB2 -z KB1 )k=m KB i+n KB j+p KB k (XIV) In the formula, i, j and k are unit vectors in the x, y and z directions of the coordinate axes, respectively; Vector S KB The modulus is the distance between the lines connecting K-B1 and K-B2, that is: Let K-B1 and Q be further defined. i The vector S of the connecting line KQ for: S KQ =(x Qi -x KB1 )i+(y Qi -y KB1 )j+(z Qi -z KB1 )k=m QK i+n QK j+p QK k (XVI) According to spatial analytic geometry, Q i The perpendicular distance from the point to the line connecting K-B1 and K-B2 is: The numerator on the right side of the above equation is the magnitude of the cross product of the two vectors, and the denominator is the result of the calculation of equation (XV). The magnitude of the above vector is calculated using the following formula: Each Q is calculated sequentially using the above formulas (XV), (XIX), and (XVII). i The depth of the wound is the vertical distance between the point and the line connecting K-B1 and K-B2, with the maximum value being the wound depth.

5. A method for constructing a three-dimensional finite element model for analyzing head injury cases according to any one of claims 1 to 4, characterized in that, in the first step, the measured data of the victim's head includes: The location and morphological parameters of the wound on the victim's skull, and the three-dimensional geometric data of the victim's skull; wherein, the morphological parameters of the wound include the length and depth of the wound, and the three-dimensional geometric data includes the length, height and width of the skull.

6. The method for constructing a three-dimensional finite element model for analyzing head injury cases according to claim 5, characterized in that, in the first step, the specific process of constructing a personalized head geometric model is as follows: The THUMS head model was imported into computer-aided engineering mesh processing software. The model was scaled 1:1 according to the three-dimensional geometric data of the victim's head. The quality of the 1D, 2D and 3D meshes was checked. If there were any non-standard mesh cells, they were pre-processed. The pre-processing process was as follows: first, the local area of ​​the head was re-meshed using the automatic mesh generation function, then the non-standard cells were deleted and a qualified mesh was regenerated. After that, the personalized geometric model of the victim's head was obtained. The specific process of creating a physical model of the skull using 3D printing technology is as follows: The brain, facial structure, and teeth of the victim's personalized skull geometry model were reverse reconstructed. The brain includes the frontal bone, parietal bone, occipital bone, temporal bone, sphenoid bone, and ethmoid bone; the facial structure includes the maxilla, mandible, zygomatic bone, lacrimal bone, nasal bone, palatine bone, inferior nasal turbinate bone, and vomer. During the reverse reconstruction, surface processing software was used to fill holes, smooth out imperfections, optimize the model, and perform three-dimensional surface reconstruction. The reconstructed model was then imported into medical image processing software to convert it into a 3D printing model. Finally, a physical model of the skull was created at a 1:1 scale using 3D printing technology.

7. The method for constructing a three-dimensional finite element model for analyzing head injury cases according to claim 6, characterized in that, The material used for printing the top part of the skull is polyetheretherketone (PEEK); the material used for printing the rest of the skull is polylactic acid (PLA).

8. A method for constructing a three-dimensional finite element model for analyzing head injury cases according to any one of claims 1 to 4, characterized in that, In the third step, the specific process of establishing a finite element model of the skull based on the personalized skull geometry model and assembling it with the finite element model of the blade includes: S1. Computer-aided engineering mesh processing software was used to divide the personalized skull geometric model into unit meshes and establish a personalized skull finite element model. During the establishment process, the layered modeling function was used to divide the skull into several layers with a thickness of 0.1 mm, and the thicknesses of the outer cortical bone, the middle cancellous bone, and the inner cortical bone were constructed to match the actual anatomical structure. The thickness of the outer cortical bone was 2.4±0.1 mm, the thickness of the middle cancellous bone was 3.2±0.1 mm, and the thickness of the inner cortical bone was 2.4±0.1 mm. S2. In computer-aided engineering mesh processing software, hexahedral elements are used to refine the mesh of the contact area between the tool and the skull, and progressive tetrahedral elements are used to process the mesh of the transition area between the contact area and the non-contact area. The main steps of the above processing include: using the mesh subdivision function to refine the mesh of the contact area between the tool and the skull, with a minimum mesh size of 0.09 mm; using the mesh refinement strategy in the transition area between the contact area and the non-contact area to divide the transition area mesh into progressive tetrahedral elements; the finite element model of the skull after the above processing includes five major tissues: outer cortical bone, middle cancellous bone, inner cortical bone, gray matter, and white matter. S3. Use a node merging strategy to handle the shared nodes of the mesh in the skull finite element model, so that the skull finite element model has consistent mechanical behavior and geometric continuity. S4. If the actual knife has been found in the case of head injury, a finite element model of the knife shall be established based on the actual knife. The main steps include: measuring the geometric parameters of the actual knife with vernier calipers; using the blade geometry data obtained by measuring the preset knife used in the second step of the simulated head injury experiment with a scanning electron microscope to reverse reconstruct the outer contour of the actual knife; establishing the knife geometric model using a CAD system; importing the model into computer-aided engineering mesh processing software for mesh generation; and performing local densification processing on the blade area. If no actual knife is found in a case of head injury, the possible types of knives in the case are pre-defined, and then the corresponding finite element model of the knife is selected from the existing knife database. S5. In the finite element software, based on the cutting angle of the tool and the contact point between the tool and the head in the second step of the simulated cutting experiment, assemble the finite element model of the tool and the finite element model of the head. S6. In the finite element software, set the parameters of each constitutive equation of the head finite element model and the parameters of the constitutive equation of the knife finite element model. The parameters include density, elastic modulus or Young's modulus, Poisson's ratio, ultimate stress, and ultimate strain. Based on the cadaver experimental data in previous literature, assign the corresponding material properties to each constitutive equation of the head finite element model. Assign the constitutive equation of the knife finite element model to a steel isotropic material.

9. A method for constructing a three-dimensional finite element model for analyzing head injury cases according to claim 8, characterized in that, The third step involves a computer simulation of the formation of a head injury, which includes the following specific processes: Computer simulation was performed using the solver of finite element software. The boundary conditions were determined by the impact position, impact angle, and impact speed obtained from the second step of the simulated impact experiment, and the bottom of the skull finite element model was fixed. In the skull finite element model, the failure mode of the bone tissue element was set to ductile fracture failure, and the failure mode of the soft tissue element was set to shear failure. During the computer simulation, when the load on the bone tissue element and soft tissue element in the contact area between the tool and the skull reached the failure condition, the element was set to detach and removed from the calculation.

10. A method for constructing a three-dimensional finite element model for analyzing head injury cases according to any one of claims 1 to 4, characterized in that, The specific process of the fourth step is as follows: Compare the wound circumference obtained from the simulation results in step 3 with the wound circumference obtained from the morphological index of the wound in step 2, and compare the wound depth obtained from the simulation results in step 3 with the wound depth obtained from the morphological index of the wound in step 2. If the difference is less than 15%, the finite element model of the skull is considered valid. Otherwise, the finite element model of the skull is calibrated and then the computer simulation step in step 3 is taken to obtain the simulation results again, and the above comparison is continued until the finite element model of the skull is considered valid. The calibration measures include adjusting the element mesh and / or adjusting the boundary conditions. Finally, the valid finite element model of the skull is taken as the final finite element model of the skull.

Images