CN-121971207-A - Artificial shoulder joint prosthesis design method and system

CN121971207ACN 121971207 ACN121971207 ACN 121971207ACN-121971207-A

Abstract

The invention discloses a method and a system for designing an artificial shoulder joint prosthesis, wherein the method comprises the steps of collecting motion tracks and muscle activation states of a shoulder joint under different motion modes, separating feature frequencies and amplitudes of shoulder joint motion, extracting dynamic motion feature sets of the shoulder joint, inputting the dynamic motion feature sets into a contact mechanics model based on a discrete element method, simulating contact force and friction behaviors between the shoulder joint prosthesis and bones to obtain contact mechanics characteristics of the prosthesis and the bones, generating an initial structure of the prosthesis by using a topology optimization algorithm based on a generation countermeasure network according to the contact mechanics characteristics, and generating a prosthesis topology structure with gradient material distribution to obtain an optimized prosthesis design scheme. With embodiments of the present invention, a prosthesis design with gradient material distribution can be created, reducing the weight of the prosthesis and enhancing its strength and durability.

Inventors

ZHAO LIANG
LIAO DONG
Qiu Gengtao
CHEN ZHENHUA
Liao Zheting
WANG ZIQI
CHEN YUFAN
WU DESHENG

Assignees

广州医科大学附属第一医院（广州呼吸中心）

Dates

Publication Date: 20260505
Application Date: 20260122

Claims (10)

1. A method of designing an artificial shoulder joint prosthesis, the method comprising: According to dynamic motion capture data and electromyographic signals of a patient, acquiring motion tracks and muscle activation states of the shoulder joint in different motion modes, separating out characteristic frequency and amplitude of the shoulder joint motion by adopting a dynamic mode decomposition algorithm based on time sequence analysis, and extracting a dynamic motion characteristic set of the shoulder joint; Inputting the dynamic motion feature set into a contact mechanical model based on a discrete element method, and simulating contact force and friction behavior between a shoulder joint prosthesis and a bone, wherein the contact mechanical model optimizes the precision of contact force distribution through a self-adaptive contact algorithm and a dynamic friction coefficient adjustment mechanism to obtain the contact mechanical characteristic of the prosthesis and the bone; Generating an initial structure of the prosthesis by using a topology optimization algorithm based on a generation countermeasure network according to the contact mechanical characteristics, wherein the topology optimization algorithm generates a prosthesis topology structure with gradient material distribution by combining material microstructure optimization and multiscale mechanical property analysis, and obtains an optimized prosthesis design scheme.
2. The method according to claim 1, wherein the capturing motion trajectories and muscle activation states of the shoulder joint in different motion modes according to the dynamic motion capturing data and the electromyographic signals of the patient, and separating feature frequencies and amplitudes of the shoulder joint motion by using a dynamic mode decomposition algorithm based on time series analysis, and extracting a dynamic motion feature set of the shoulder joint, comprises: According to dynamic motion capture data and electromyographic signals of a patient, a data acquisition frame based on edge calculation is adopted to acquire the motion trail and the muscle activation state of the shoulder joint in different motion modes in real time, and the real-time performance and the continuity of data acquisition are ensured through a lightweight data caching mechanism; the motion track signal and the muscle activation state signal are subjected to differential filtering processing aiming at noise characteristics of different data sources through a dynamic threshold adjustment mechanism by adopting a noise filtering algorithm based on wavelet transformation to separate environmental noise and interference in the signals, so as to generate a preliminary standardized data set; For the preliminary standardized data set, a dynamic mode decomposition algorithm based on time sequence analysis is adopted to separate the characteristic frequency and amplitude of shoulder joint movement, and key characteristics under different movement modes are captured through a self-adaptive decomposition layer number selection mechanism to generate preliminary dynamic movement characteristic representation; And carrying out weighted fusion on the motion trail characteristics and the muscle activation characteristics by adopting a characteristic fusion-based method for the preliminary dynamic motion characteristic representation, capturing the association relationship between the motion trail and the muscle activation through a cross-modal attention mechanism, and generating a final shoulder joint dynamic motion characteristic set.
3. The method according to claim 2, wherein the inputting the dynamic motion feature set into a contact mechanics model based on a discrete element method simulates contact force and friction behavior between a shoulder joint prosthesis and a bone, wherein the contact mechanics model optimizes precision of contact force distribution through an adaptive contact algorithm and a dynamic friction coefficient adjustment mechanism to obtain contact mechanics characteristics of the prosthesis and the bone, and comprises: according to the three-dimensional geometric model of the shoulder joint prosthesis and the bone, the prosthesis and the bone are discretized into discrete meta-particles by adopting a modeling technology based on a discrete meta-method, and a contact area is locally encrypted by a self-adaptive particle division algorithm to generate a high-precision discrete meta-model; for the discrete element model, a contact force calculation method based on a self-adaptive contact algorithm is adopted, a motion track and a muscle activation state in a dynamic motion feature set are combined, contact force distribution between a prosthesis and bones is simulated, and a friction coefficient is dynamically adjusted according to a motion mode and contact pressure through a dynamic friction coefficient adjustment mechanism, so that preliminary contact mechanical characteristics are generated; For the initial contact mechanical characteristics, adopting a contact force distribution optimization method based on iterative optimization, combining the material properties of the prosthesis and the mechanical characteristics of bones, optimizing the precision of the contact force distribution, dynamically adjusting the parameters of a contact algorithm through an error feedback mechanism, and generating optimized contact force distribution; And (3) for the optimized contact force distribution, a contact mechanical property generation method based on a visualization technology is adopted to map the contact force and friction behavior onto a three-dimensional geometric model, and an intuitive prosthesis and bone contact mechanical property diagram is generated through a color gradient mapping and contour drawing technology.
4. A method according to claim 3, wherein the generating an initial structure of the prosthesis from the contact mechanics characteristics using a topology optimization algorithm based on generating a countermeasure network, wherein the topology optimization algorithm generates a prosthesis topology with gradient material distribution by combining material microstructure optimization and multiscale mechanical properties analysis, resulting in an optimized prosthesis design scheme, comprising: According to the contact mechanical characteristics, adopting a topology optimization algorithm based on a generation countermeasure network, combining the mechanical requirement and the biocompatibility requirement of the prosthesis, generating an initial prosthesis structure, ensuring that the generated structure meets the optimization targets of contact force distribution and friction behavior through a condition constraint mechanism, and generating a preliminary prosthesis topology structure; For the preliminary prosthesis topological structure, a material microstructure-based optimization method is adopted, the microstructure of the material is optimized by combining the mechanical property and biocompatibility indexes of the prosthesis, and the influence of the microstructure on the macroscopic mechanical property is captured through multi-scale mechanical property analysis, so that the preliminary optimized material distribution is generated; for the primarily optimized material distribution, adopting a topological optimization method based on gradient material distribution, combining stress distribution and friction behavior in contact mechanical characteristics, dynamically adjusting the gradient distribution of the material, and generating a prosthesis topological structure with gradient material distribution through a self-adaptive gradient adjustment mechanism; and (3) for the prosthesis topological structure with gradient material distribution, adopting a verification method based on finite element analysis, combining the contact mechanical property and the patient anatomical model, verifying the mechanical property and biocompatibility of the design scheme, and dynamically adjusting the design parameters through a feedback correction mechanism to generate the final optimized prosthesis design scheme.
5. A prosthetic shoulder joint design system, the system comprising: The acquisition module is used for acquiring the motion trail and the muscle activation state of the shoulder joint under different motion modes according to the dynamic motion capture data and the electromyographic signals of the patient, separating out the characteristic frequency and the amplitude of the shoulder joint motion by adopting a dynamic mode decomposition algorithm based on time sequence analysis, and extracting the dynamic motion characteristic set of the shoulder joint; The simulation module is used for inputting the dynamic motion feature set into a contact mechanical model based on a discrete element method and simulating the contact force and friction behavior between the shoulder joint prosthesis and the bone, wherein the contact mechanical model optimizes the precision of contact force distribution through a self-adaptive contact algorithm and a dynamic friction coefficient adjustment mechanism to obtain the contact mechanical characteristic of the prosthesis and the bone; the generation module is used for generating an initial structure of the prosthesis by utilizing a topology optimization algorithm based on a generation countermeasure network according to the contact mechanical characteristics, wherein the topology optimization algorithm is used for generating a prosthesis topology structure with gradient material distribution by combining material microstructure optimization and multiscale mechanical property analysis, and an optimized prosthesis design scheme is obtained.
6. The system according to claim 5, wherein the acquisition module is specifically configured to: According to dynamic motion capture data and electromyographic signals of a patient, a data acquisition frame based on edge calculation is adopted to acquire the motion trail and the muscle activation state of the shoulder joint in different motion modes in real time, and the real-time performance and the continuity of data acquisition are ensured through a lightweight data caching mechanism; the motion track signal and the muscle activation state signal are subjected to differential filtering processing aiming at noise characteristics of different data sources through a dynamic threshold adjustment mechanism by adopting a noise filtering algorithm based on wavelet transformation to separate environmental noise and interference in the signals, so as to generate a preliminary standardized data set; For the preliminary standardized data set, a dynamic mode decomposition algorithm based on time sequence analysis is adopted to separate the characteristic frequency and amplitude of shoulder joint movement, and key characteristics under different movement modes are captured through a self-adaptive decomposition layer number selection mechanism to generate preliminary dynamic movement characteristic representation; And carrying out weighted fusion on the motion trail characteristics and the muscle activation characteristics by adopting a characteristic fusion-based method for the preliminary dynamic motion characteristic representation, capturing the association relationship between the motion trail and the muscle activation through a cross-modal attention mechanism, and generating a final shoulder joint dynamic motion characteristic set.
7. The system according to claim 6, characterized in that said simulation module is in particular adapted to: according to the three-dimensional geometric model of the shoulder joint prosthesis and the bone, the prosthesis and the bone are discretized into discrete meta-particles by adopting a modeling technology based on a discrete meta-method, and a contact area is locally encrypted by a self-adaptive particle division algorithm to generate a high-precision discrete meta-model; for the discrete element model, a contact force calculation method based on a self-adaptive contact algorithm is adopted, a motion track and a muscle activation state in a dynamic motion feature set are combined, contact force distribution between a prosthesis and bones is simulated, and a friction coefficient is dynamically adjusted according to a motion mode and contact pressure through a dynamic friction coefficient adjustment mechanism, so that preliminary contact mechanical characteristics are generated; For the initial contact mechanical characteristics, adopting a contact force distribution optimization method based on iterative optimization, combining the material properties of the prosthesis and the mechanical characteristics of bones, optimizing the precision of the contact force distribution, dynamically adjusting the parameters of a contact algorithm through an error feedback mechanism, and generating optimized contact force distribution; And (3) for the optimized contact force distribution, a contact mechanical property generation method based on a visualization technology is adopted to map the contact force and friction behavior onto a three-dimensional geometric model, and an intuitive prosthesis and bone contact mechanical property diagram is generated through a color gradient mapping and contour drawing technology.
8. The system according to claim 7, wherein the generating module is specifically configured to: According to the contact mechanical characteristics, adopting a topology optimization algorithm based on a generation countermeasure network, combining the mechanical requirement and the biocompatibility requirement of the prosthesis, generating an initial prosthesis structure, ensuring that the generated structure meets the optimization targets of contact force distribution and friction behavior through a condition constraint mechanism, and generating a preliminary prosthesis topology structure; For the preliminary prosthesis topological structure, a material microstructure-based optimization method is adopted, the microstructure of the material is optimized by combining the mechanical property and biocompatibility indexes of the prosthesis, and the influence of the microstructure on the macroscopic mechanical property is captured through multi-scale mechanical property analysis, so that the preliminary optimized material distribution is generated; for the primarily optimized material distribution, adopting a topological optimization method based on gradient material distribution, combining stress distribution and friction behavior in contact mechanical characteristics, dynamically adjusting the gradient distribution of the material, and generating a prosthesis topological structure with gradient material distribution through a self-adaptive gradient adjustment mechanism; and (3) for the prosthesis topological structure with gradient material distribution, adopting a verification method based on finite element analysis, combining the contact mechanical property and the patient anatomical model, verifying the mechanical property and biocompatibility of the design scheme, and dynamically adjusting the design parameters through a feedback correction mechanism to generate the final optimized prosthesis design scheme.
9. A storage medium having a computer program stored therein, wherein the computer program is arranged to perform the method of any of claims 1-4 when run.
10. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, the processor being arranged to run the computer program to perform the method of any of claims 1-4.

Description

Artificial shoulder joint prosthesis design method and system Technical Field The invention belongs to the technical field of design, and particularly relates to a method and a system for designing an artificial shoulder joint prosthesis. Background With the advancement of medical technology and the improvement of living standard of people, shoulder joint replacement surgery has become one of the effective methods for treating shoulder joint diseases (such as shoulder osteoarthritis, rotator cuff injury, etc.). The design and manufacture of artificial shoulder prostheses involves a number of disciplinary knowledge including biomechanics, materials science, computer science, and medical engineering, among others. However, the current artificial shoulder joint prosthesis design method still has some defects, which are mainly characterized by insufficient consideration of individual differences of patients, inaccurate simulation of interactions between the prosthesis and the biological skeleton, and lack of flexibility and intelligence in the design process. Disclosure of Invention The invention aims to provide a method and a system for designing an artificial shoulder joint prosthesis, which are used for solving the defects in the prior art, and can generate a prosthesis design scheme with gradient material distribution, reduce the weight of the prosthesis and enhance the strength and the durability of the prosthesis. One embodiment of the present application provides a method of designing an artificial shoulder joint prosthesis, the method comprising: According to dynamic motion capture data and electromyographic signals of a patient, acquiring motion tracks and muscle activation states of the shoulder joint in different motion modes, separating out characteristic frequency and amplitude of the shoulder joint motion by adopting a dynamic mode decomposition algorithm based on time sequence analysis, and extracting a dynamic motion characteristic set of the shoulder joint; Inputting the dynamic motion feature set into a contact mechanical model based on a discrete element method, and simulating contact force and friction behavior between a shoulder joint prosthesis and a bone, wherein the contact mechanical model optimizes the precision of contact force distribution through a self-adaptive contact algorithm and a dynamic friction coefficient adjustment mechanism to obtain the contact mechanical characteristic of the prosthesis and the bone; Generating an initial structure of the prosthesis by using a topology optimization algorithm based on a generation countermeasure network according to the contact mechanical characteristics, wherein the topology optimization algorithm generates a prosthesis topology structure with gradient material distribution by combining material microstructure optimization and multiscale mechanical property analysis, and obtains an optimized prosthesis design scheme. Optionally, the capturing data and electromyographic signals according to the dynamic motion of the patient collect the motion track and muscle activation state of the shoulder joint in different motion modes, and adopt a dynamic mode decomposition algorithm based on time sequence analysis to separate the characteristic frequency and amplitude of the shoulder joint motion, and extract the dynamic motion characteristic set of the shoulder joint, including: According to dynamic motion capture data and electromyographic signals of a patient, a data acquisition frame based on edge calculation is adopted to acquire the motion trail and the muscle activation state of the shoulder joint in different motion modes in real time, and the real-time performance and the continuity of data acquisition are ensured through a lightweight data caching mechanism; the motion track signal and the muscle activation state signal are subjected to differential filtering processing aiming at noise characteristics of different data sources through a dynamic threshold adjustment mechanism by adopting a noise filtering algorithm based on wavelet transformation to separate environmental noise and interference in the signals, so as to generate a preliminary standardized data set; For the preliminary standardized data set, a dynamic mode decomposition algorithm based on time sequence analysis is adopted to separate the characteristic frequency and amplitude of shoulder joint movement, and key characteristics under different movement modes are captured through a self-adaptive decomposition layer number selection mechanism to generate preliminary dynamic movement characteristic representation; And carrying out weighted fusion on the motion trail characteristics and the muscle activation characteristics by adopting a characteristic fusion-based method for the preliminary dynamic motion characteristic representation, capturing the association relationship between the motion trail and the muscle activation through a cross-modal attention mechanism, and generati