Digital Twins in Neurology Care: Evidence, Limitations, and a Path Forward

Area of Science:

• Neurology
• Artificial Intelligence
• Computational Medicine

Background:

• Digital twins are increasingly promoted in neurology, yet often lack conceptual and methodological rigor.
• Strict definitions require dynamically updated, bidirectionally linked models with predictive value, criteria rarely met by current neurological applications.

Purpose of the Study:

• Critically examine the current state of digital twin applications across major neurological domains.
• Differentiate between true clinical digital twins and twin-inspired models in neurological practice.

Main Methods:

• Critical review of digital twin applications in dementia, multiple sclerosis, Parkinson's disease, epilepsy, stroke, pain, and migraine.
• Analysis based on established criteria for digital twin conceptualization and methodological rigor.

Main Results:

• Most existing neurological systems function as twin-inspired longitudinal decision-support or trial-analytics models, not true clinical digital twins.
• Progress in neurological digital twins is hindered by issues in measurement validity, uncertainty handling, prospective evaluation, and governance.

Conclusions:

• Neurology is well-suited for digital twin approaches due to disease complexity and data richness.
• A pragmatic path forward involves developing question-specific, validated neurological digital twins rather than generalized brain twin concepts.