From Skull Density Ratio‐based eligibility to efficiency‐informed planning: Predictive modeling of thermal response in MR‐guided focused ultrasound

Background Transcranial MR‐guided focused ultrasound (MRgFUS) enables incisionless thermoablation of deep brain targets and has become an established modality in functional neurosurgery. Treatment efficiency is strongly influenced by skull‐mediated ultrasound attenuation. Although the skull density ratio (SDR) is currently the primary parameter used for pre‐procedural screening, it provides only a partial description of skull‐related variability in thermal response and offers limited guidance for intra‐procedural energy titration. Purpose To comprehensively characterize skull‐related determinants of thermal efficiency beyond SDR and to develop multivariate models for predicting treatment efficiency and sonication‐wise peak temperature to support patient screening and intra‐procedural decision‐making. Methods We retrospectively analyzed 316 MRgFUS thermoablative procedures (214 thalamotomies and 102 subthalamotomies). Skull metrics derived from CT included SDR, skull thickness (ST), diploe thickness (DT), angle of incidence (AOI), and higher‐order distributional descriptors. Thermal efficiency was quantified using multiple temperature‐energy‐based estimators. Two predictive models were developed, each validated on an independent held‐out test set (20% of procedures): (1) a forward‐selected multivariate linear regression model for procedure‐level thermal efficiency (temperature‐to‐energy ratio at 55°C‐TER 55 ), guided by adjusted R 2 and AIC, with performance reported as R 2 ; and (2) a gradient boosting model for sonication‐wise peak temperature prediction, integrating sonication parameters, skull metrics, and cumulative treatment history, with performance reported as MAE globally and stratified by temperature range. Results SDR, ST, and DT were significantly associated with thermal efficiency, and composite metrics such as SDR/DT outperformed SDR alone. A five‐feature multivariable linear regression model based on skull‐derived metrics achieved an adjusted R 2 of 0.63 (cross‐validated R 2 = 0.61, MAE = 0.54°C/kJ) for predicting TER 55 , with consistent generalization on an independent held‐out test set ( R 2 = 0.58, MAE = 0.53°C/kJ). The peak temperature prediction model achieved a cross‐validated MAE of 1.70°C on the training set and 1.91°C on a fully held‐out procedure‐level test set, with stable performance across clinically dominant temperature ranges (< 60°C). Prediction accuracy was reduced at ≥60°C, reflecting data imbalance and systematic underestimation at extreme temperatures. Conclusions Thermal efficiency in MRgFUS cannot be adequately characterized by SDR alone. A multivariate efficiency model integrating five skull descriptors substantially outperforms SDR‐based screening, providing a more informative and continuous representation of treatment feasibility and enabling pre‐procedural estimation of expected energy requirements. When combined with treatment parameters and cumulative thermal history, multivariate models enable accurate sonication‐wise temperature prediction, supporting efficiency‐informed energy titration and dynamic procedural guidance, and advancing MRgFUS toward more precise and personalized therapy.