Amortizing personalization in virtual brain twins

TL;DR

This paper introduces a method called anonymized personalization for virtual brain twins, enabling personalized, resource-efficient inference while maintaining data privacy, with demonstrated reliability and potential impact on neuroscience.

Contribution

It proposes a novel amortized inference approach that allows personalized brain models to be trained anonymously and inferred efficiently, addressing privacy and resource challenges.

Findings

Demonstrated reliable personalized inference in a case study.

Enabled lightweight inference suitable for clinical applications.

Provided open-source code for implementation.

Abstract

Virtual brain twins are personalized digital models of individual human subject or patient's brains, allowing for mechanistic interpretation of neuroimaging data features. Training and inference with these models however presents a pair of challenges: large shared infrastructure do not allow for use of personal data and inference in clinical applications should not require significant resources. We introduce "anonymized personalization" to address both by expanding model priors to include personalization which under amortized inference allows training to be performed anonymously, while inference is both personalized and lightweight. We illustrate the basic approach, demonstrate reliability in an example, and discuss the impact on both experimental and computational neuroscience. Code is available at https://github.com/ins-amu/apvbt.