Ground-truth validation of uni- and multivariate lesion inference approaches

Abstract

AbstractLesion analysis reveals causal contributions of brain regions to brain functions. Various strategies have been used for inferring brain functions from brain lesions. These approaches can be broadly categorized as univariate or multivariate methods. Here we analysed data from 581 patients with acute ischemic injury, parcellated into 41 Brodmann areas, and systematically investigated the inferences made by univariate and multivariate lesion analysis methods via ground-truth simulations, in which we defined a priori contributions of brain areas to assumed brain function. Particularly, we analysed single-region models, with only single areas presumed to contribute functionally, and multiple-region models, with two contributing regions. The analyses consistently showed a considerably better performance of multivariate than univariate methods in terms of accuracy and mis-inference error. Specifically, the univariate approaches of Lesion Symptom Mapping as well as Lesion Symptom Correlation mis-inferred substantial contributions from several areas even in the single-region models, and also when accounting for lesion size. Of the multivariate approaches, the game-theoretical Multi-perturbation Shapley value Analysis typically showed the best performance. Our findings suggest that multivariate approaches produce highly reliable lesion inferences, without requiring lesion size consideration, while the application of univariate methods may yield substantial mis-localizations that limit the reliability of functional attributions.