Virtual brain models: a roadmap towards effective brain digital twins
5 affiliations
- 1Department of Brain and Behavioral Sciences, University of Pavia, Pavia, Italy
- 2Digital Neuroscience Center, IRCCS Mondino Foundation, Pavia, Italy
- 3NMR Research Unit, Queen Square MS Centre, Department of Neuroinflammation, UCL Queen Square Institute of Neurology, Faculty of Brain Sciences, UCL (University College London), London, United Kingdom
- 4CNRS, Paris-Saclay Institute of Neuroscience (NeuroPSI), Paris-Saclay University, Gif-sur-Yvette, France
- 5Institut de Neurosciences des Systèmes, INSERM, INS, Aix Marseille University, Marseille, France
Abstract
Virtual Brain Models (VBMs) are emerging computational tools promising to revolutionize personalised and precision medicine. VBMs reproduce brain functional dynamics through a generative model that can be constructed and optimized against single subject data, thereby generating what is called a brain digital twin (BDT). VBMs can provide critical information about parameters of network functioning and connectivity but this ability depends on their design, raising some fundamental questions. Can VBM parameters identify key physiological features of the brain otherwise not accessible? Are VBM parameters correlated with sensorimotor and cognitive functions? What is still needed to make a VBM more precise in its physiological mechanisms? Simulations in patients affected by neurological diseases are indeed showing that VBMs can operate as effective BDTs capturing essential aspects of the physiopathology of single patients along with relevant physio-psychological correlations. However, the current VBM theory and technology still pose some limitations. We maintain that the evolution of VBM holds the key for a deeper understanding of brain physiology and pathology fostering the development of powerful computational tools capable of supporting personalized clinical decision making. Here, we will consider a roadmap addressing what VBM already do well, why this is not enough, what must change, and what becomes possible if it does.
Related
- Multiscale predictive cellular modeling: integrating hypothesis grammars, digital twins, and multi-omics for In silico oncology and precision theranostics
- OmniBioTwin: A System-of-Twinned-Systems Framework for Health Digital Twins
- An updated roadmap for brain-mental health associations in the UK Biobank
- Brain Initiative Roadmap: Building the Next Decade of NeuroAI
- APOD: reasoning-guided agentic population ordinary differential equation discovery for pharmacological digital twins
BibTeX
@article{dangelo2026virtual,
title = {{Virtual brain models: a roadmap towards effective brain digital twins}},
author = {Egidio D’Angelo and Claudia A. M. Gandini Wheeler-Kingshott and Alain Destexhe and Viktor Jirsa},
journal = {Frontiers in Computational Neuroscience},
year = {2026},
month = oct,
doi = {10.3389/fncom.2026.1973956},
url = {https://doi.org/10.3389/fncom.2026.1973956},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Virtual brain models: a roadmap towards effective brain digital twins
AU - Egidio D’Angelo
AU - Claudia A. M. Gandini Wheeler-Kingshott
AU - Alain Destexhe
AU - Viktor Jirsa
PY - 2026
DA - 2026/10/29
JO - Frontiers in Computational Neuroscience
DO - 10.3389/fncom.2026.1973956
UR - https://doi.org/10.3389/fncom.2026.1973956
AB - Virtual Brain Models (VBMs) are emerging computational tools promising to revolutionize personalised and precision medicine. VBMs reproduce brain functional dynamics through a generative model that can be constructed and optimized against single subject data, thereby generating what is called a brain digital twin (BDT). VBMs can provide critical information about parameters of network functioning and connectivity but this ability depends on their design, raising some fundamental questions. Can VBM parameters identify key physiological features of the brain otherwise not accessible? Are VBM parameters correlated with sensorimotor and cognitive functions? What is still needed to make a VBM more precise in its physiological mechanisms? Simulations in patients affected by neurological diseases are indeed showing that VBMs can operate as effective BDTs capturing essential aspects of the physiopathology of single patients along with relevant physio-psychological correlations. However, the current VBM theory and technology still pose some limitations. We maintain that the evolution of VBM holds the key for a deeper understanding of brain physiology and pathology fostering the development of powerful computational tools capable of supporting personalized clinical decision making. Here, we will consider a roadmap addressing what VBM already do well, why this is not enough, what must change, and what becomes possible if it does.
SN - 1662-5188
ER -