Real-time simulation for multi-component biomechanical analysis using localized tissue constraint progressive...

Area of Science:

• Biomedical Engineering
• Computer Science
• Surgical Simulation

Background:

• Accurate real-time simulation of organ deformation is vital for virtual surgical training.
• Anisotropic and nonlinear organ characteristics pose challenges for high-fidelity simulation.

Purpose of the Study:

• To develop an efficient and accurate method for simulating multi-component tissue deformation in virtual surgery.
• To address the limitations of existing models in representing complex biomechanical properties.

Main Methods:

• A localized tissue constraint progressive transfer learning method was proposed.
• Developed a base-compensated dual-output transfer learning strategy and a progressive learning architecture.
• Enriched biomechanical datasets and utilized focused models for varied tissue properties.

Main Results:

• The proposed method achieved 50% higher accuracy compared to four state-of-the-art methods.
• Stable simulation of organ deformations was achieved at 0.005 seconds per frame.
• Demonstrated significant improvement in computational efficiency.

Conclusions:

• The localized tissue constraint progressive transfer learning method offers superior accuracy and efficiency for virtual surgical training.
• This approach enables more realistic simulation of complex organ biomechanics.
• The method effectively handles multi-component tissue properties, advancing the field of surgical simulation.