Uncertainty mapping and probabilistic tractography using Simulation-based Inference in diffusion MRI: A comparison with classical Bayes

J. P. Manzano-Patrón, Michael Deistler, Cornelius Schröder, Theodore Kypraios, Pedro J. Gonçalves, Jakob H. Macke, Stamatios N. Sotiropoulos

Abstract: Highlights•We present a novel Simulation-Based Inference framework for brain diffusion MRI.•We show its use for mapping uncertainty and performing probabilistic tractography.•An extensive comparison is conducted against established Bayesian methods (MCMC).•Amortised SBI of model parameter posterior is as accurate and remarkably faster.•SBI-based tractography agreement with MCMC is within scan-rescan reproducibility.