A Novel Nonlinear Parameter Estimation Method of Soft Tissues

Area of Science:

• Biomechanics
• Computational mechanics
• Medical imaging

Background:

• Accurate elastic parameters of soft tissues are crucial for medical diagnosis and virtual surgery simulations.
• Existing methods for parameter estimation can be complex and prone to local minima.
• Precise force and deformation data are essential for reliable tissue modeling.

Purpose of the Study:

• To develop a novel nonlinear parameter estimation method for soft tissues.
• To enhance the accuracy of force and deformation measurements for tissue characterization.
• To improve the robustness and precision of soft tissue elastic parameter estimation.

Main Methods:

• Utilized an in-house data acquisition platform for force and deformation measurements.
• Employed a weighted combination forecasting model based on support vector machine (WCFM_SVM) for force correction.
• Developed a tetrahedral finite element parameter estimation model using Young's modulus and Poisson's ratio substitution.
• Incorporated initial parameters from a linear finite element model to enhance robustness.
• Implemented a self-adapting Levenberg-Marquardt (LM) algorithm for parameter estimation.

Main Results:

• The WCFM_SVM model achieved a maximum absolute error of less than 0.03 Newton for force correction, outperforming other models.
• The parameter estimation model demonstrated a maximum absolute error of less than 1.5 mm between calculated and measured nodal displacements.
• The proposed method yielded precise nonlinear parameters for soft tissues.

Conclusions:

• The novel nonlinear parameter estimation method significantly improves the accuracy of soft tissue characterization.
• The WCFM_SVM and self-adapting LM algorithm contribute to robust and precise estimation of elastic parameters.
• This method holds potential for advancing medical diagnosis and virtual surgery simulation accuracy.