Anatomical correlates of delivered TIS fields across deep and cortical targets

The objective of this study was to quantify inter-individual variability in target-region TIS envelope amplitude and to examine its associations with anatomical features across deep and cortical targets.

We built individualized finite element head models for 502 healthy HCP participants aged 22-35 years and evaluated five target ROI-montage pairs: hippocampus (two montages), left and right thalamus, and motor cortex. For each target ROI-montage pair, we quantified mean TIS envelope amplitude within target region (EA mean , V/m) and the target-to-brain envelope-amplitude ratio (TBR) as measures of target-field magnitude and spatial selectivity, respectively. Two prespecified global volume ratios were assessed with Pearson correlations, and 20 anatomical features were evaluated using two-stage LASSO-OLS models with an 80:20 internal hold-out split. Bootstrap selection frequencies were calculated for the EA mean models.

EA mean showed substantial inter-individual variability (CV = 28.1-40.0%). Brain-TIV ratio and GM-head ratio were positively correlated with EA mean across all five target ROI-montage pairs (r = 0.281-0.420; all p < 0.001), with directions compatible with CSF-related current shunting. Pairwise EA mean correlations across the five target ROI-montage pairs were high (r = 0.749-0.957). Five two-stage LASSO-OLS models yielded in-sample R 2 values of 0.185-0.282. For TBR, MC1 showed the strongest model fit (R 2 = 0.311; hold-out R 2 = 0.267), while no predictor reached p < 0.05 in either thalamic model. Global predictors appeared repeatedly in the deep-target models. In MC1, the significant set also included skull-head ratio and W-scalp thickness (MC1), neither of which was significant in the deep-target models. The two hippocampal montages (H1 and H2) differed by 31% in EA mean , and their predictor sets reaching OLS p < 0.05 showed modest overlap (Jaccard = 0.33).

This population-scale computational dosimetry analysis quantifies how fixed TIS currents and montages translate into target-field magnitudes across individual head models. The results distinguish the anatomical correlates of delivered amplitude from those of spatial selectivity and provide quantitative information for anatomy-informed TIS planning.