Real-time tumor ablation simulation based on the dynamic mode decomposition method.

Area of Science:

• Biomedical Engineering
• Computational Biology
• Medical Physics

Background:

• Accurate real-time simulation of tumor ablation is crucial for effective medical treatment planning.
• Existing methods often involve complex boundary treatments and approximations.
• The Pennes bioheat model is a standard for simulating heat transfer in biological tissues.

Purpose of the Study:

• To develop a reliable, real-time forecasting method for tumor ablation simulations using Dynamic Mode Decomposition (DMD).
• To incorporate water evaporation and tissue damage into an extended Pennes bioheat model for enhanced accuracy.
• To validate the DMD approach against established numerical and analytical methods.

Main Methods:

• Utilized a meshless point collocation solver for numerical solutions of the bioheat transfer equations.
• Applied the Dynamic Mode Decomposition (DMD) method to accelerate the forecasting of simulation results.
• Validated the DMD approach with simple problems and applied it to 3D tumor ablation simulations with realistic parameters.

Main Results:

• The DMD method provides highly accelerated numerical solutions for bioheat transfer, crucial for clinical applications.
• The approach accurately predicts temperature profiles in tumors and surrounding healthy tissue, even with non-linear thermal properties.
• It bypasses the complex and approximate mathematical treatment of tissue boundaries.

Conclusions:

• The low computational cost of DMD makes it suitable for in situ, real-time tumor ablation simulations without compromising accuracy.
• Tumor ablation treatment planning becomes feasible on personal computers due to the method's simplicity.
• Geometrical data from standard medical imaging modalities can be directly integrated into the simulation process.