Feasibility of thermal dose painting during magnetic nanoparticle thermal therapy: a computational analysis

Background: Spatially confined magnetic nanoparticle thermal therapy (SC-MNTT) can improve the tumor coverage index (CI) when the magnetic field-free region (FFR) is manually steered. However, static steering demands significant time and expertise from experienced control engineers during treatment planning. Objective: This study presents a computational framework for automated, dynamic FFR steering combined with real-time tumor boundary temperature control to enhance magnetic nanoparticle thermal therapy (MNTT) for cancer therapy. The framework also supports controlled MNTT treatment planning. Method: We integrated the Pennes bioheat transfer equation with a proportional-integral-derivative (PID) controller to maintain the maximum tumor boundary temperature at 46 °C. The magnetic nanoparticle (MNP) heat source was divided into four subdomains to simulate SC-MNTT. A fuzzy logic controller automatically selected the FFR based on four-point temperature measurements and corresponding thermal dose, aiming to maximize CI. We examined the influence of key treatment parameters: maximum treatment temperature (60 or 80 °C), MNP distribution (uniform or Gaussian), sensor noise, and MNP position relative to the tumor center. Results: Automated dynamic FFR steering increased CI by ∼25–35%, depending on these parameters. The highest CI occurred at a maximum treatment temperature of 80 °C with a uniform MNP distribution. Sensor noise had minimal effect, while minimal offset between the MNP position and tumor center was essential for maximizing CI. Conclusion: The proposed fuzzy logic-based dynamic FFR steering may reduce reliance on expert control engineers for SC-MNTT treatment planning. Future work will focus on pre-clinical implementation and experimental validation of the control system.