Direct-Necrosis-Monitoring-Based Adaptive Model Predictive Control for Ablation Therapy Including Patient-Specific...

Area of Science:

• Biomedical Engineering
• Therapeutic Technologies
• Control Systems

Background:

• Thermal ablation is a key treatment for difficult-to-resect tumors like glioblastoma.
• Necrosis feedback (NFB) offers accurate monitoring, surpassing traditional temperature-based methods.
• Existing NFB methods often ignore residual heat, impacting ablation outcomes.

Purpose of the Study:

• To develop an adaptive control framework (DNaMPC) for thermal ablation.
• To address challenges of residual heat management and patient-specific parameter identification.
• To improve ablation accuracy and safety using real-time necrosis monitoring.

Main Methods:

• Proposed a Direct Necrosis-Monitoring-based Adaptive Model Predictive Control (DNaMPC) framework.
• Integrated real-time NFB with residual heat management using adaptive MPC.
• Employed a hierarchical Extended Kalman Filter (EKF) for adaptive parameter estimation, focusing on the thermal damage threshold.

Main Results:

• DNaMPC eliminated excessive post-ablation heating seen with simpler controls.
• Outperformed existing methods in targeting accuracy and energy efficiency.
• Demonstrated robustness against significant variations in thermal parameters and sensing noise.

Conclusions:

• DNaMPC effectively utilizes residual heat and adapts to patient variability for precise thermal ablation.
• The framework shows promise for developing individualized and safer thermal ablation therapies.
• This approach lays the groundwork for advanced, adaptive cancer treatments.