The Study On The Identification Of Material Parameters Of Human External Nose Based On Mechanical Analysis And Machine Learning

Rhinoplasty is a plastic surgery procedure in which a prosthesis is implanted under the skin of the nose to make the nose look more beautiful.For a long time,the determination of prosthesis size in rhinoplasty mostly depended on the surgeon’s subjective experience and manual carving during operation.Improper size of the implant often leads to complications such as exposure of the prosthesis,tilt of the implant and soft tissue ulceration after rhinoplasty.Due to the lack of a quantitative method to measure the mechanical properties of the nose of the recipient,little researchers have yet determined the size of the prosthesis and discussed the causes of complications from the perspective of mechanical analysis.The purpose of this paper is to find a method to estimate the size of prosthesis in rhinoplasty based on mechanical analysis,and to provide reference for surgeons to select prosthesis before operation.The main work of this paper is as follows.(1)A nasal tension tester was designed and implemented to perform tensile tests on the nasal part of the patient and record the test data.(2)A parametric automatic modeling method is designed,which can automatically build a geometric model based on the morphological data of the nose input by the user,such as the dorsal length of the nose and the width of the nose.(3)A method for material parameter identification using the difference method is proposed.The main idea is that,for a specific nasal geometry model,several sets of loads are determined;the variance between the predicted and true displacements of the material parameters under these loadings is calculated as the objective function.Starting from an initial set of material parameters,the predicted displacements of that set are calculated,and the partial derivatives of each material parameter for each displacement are calculated using the difference method,and the material parameters are fine-tuned from this matrix of partial derivatives.Then the predicted displacement is calculated from the adjusted material parameters,the partial derivative matrix is calculated,and the material parameters are fine-tuned.So on and so forth until the objective function is less than the accuracy requirement.(4)A method for material parameter identification using artificial neural network is proposed.The parameter identification method based on the difference method has high accuracy,but it is sensitive to the selection of initial parameters,and the inappropriate selection of initial parameters will lead to non-convergence of the calculation and failure to obtain results.As a complement and contrast,the idea of the artificial neural network-based parameter identification method is to select several sets of loads and several sets of material parameters,calculate the displacement of each set of material parameters under each load,and use the displacement and material parameters as the training set.The neural network is trained to find a mapping relationship from the displacements to the material parameters.The experimental displacements are used as input,and the material parameters corresponding to the experimental displacements are calculated using the trained neural network.(5)A safety criterion for tensile deformation of the nose is proposed,and a method for calculating the upper limit of rhinoplasty prosthesis thickness is given on the basis of parameter identification and safety criterion.(6)In this paper,five volunteers were invited to perform tensile tests using a nasal tension tester,and a geometric model was established based on the test data,material parameters of the nose were identified,and the safety range of nasal tension was finally calculated for each volunteer.It was verified that the nasal tension tester designed in this paper is safe and the parameter identification method proposed in this paper is effective.The main innovations of this paper are as follows.(1)A nasal tension tester was designed and implemented.(2)Two methods for identifying the parameters of nasal materials are proposed.(3)The calculation method of the upper limit of the thickness of the prosthesis is given from the mechanical point of view.