The hot-to-cold spot quotient for SAR-based treatment planning in deep microwave hyperthermia

Abstract

Background A necessary precondition for a successful microwave hyperthermia (HT) treatment delivered by phased arrays is the ability of the HT applicator to selectively raise the temperature of the entire tumor volume. SAR-based treatment plan (HTP) optimization methods exploit the correlation between specific absorption rate (SAR) and temperature increase in order to determine the set of steering parameters for optimal focusing, while allowing for lower model complexity. Several cost functions have been suggested in the past for this optimization problem. However, their correlation with high and homogeneous tumor temperatures remains sub-optimal in many cases. Previously, we proposed the hot-to-cold spot quotient (HCQ) as a novel cost function for SAR-based HTP optimization and showed its potential to address these issues.Materials and methods In this work, we validate the HCQ on a standard ESHO patient repository within single and multi-frequency contexts. We verify its correlation with clinical SAR and temperature indexes, and compare it to HTPs obtained using a commonly accepted cost-function for SAR-based HTP (hot-spot to target quotient, HTQ).Results and discussion The results show that low HCQ values produce better SAR (TC50, TC75) and temperature metrics (T50, T90) than HTQ in most patient models and frequency settings. For the deep-seated tumors, the correlation between the clinical indicators and 1/HCQ is more favorable than the correlation exhibited by 1/HTQ.Conclusion The validation confirms the ability of HCQ to promote target coverage and hot-spot suppression in SAR-based HTP optimization, resulting in higher SAR and temperature indexes for deep-seated tumors.

Keywords

• Hyperthermia
• computational modeling
• treatment planning
• cost function
• control systems engineering/treatment optimization
• SAR