Traumatic brain injury

Traumatic brain injury (TBI) is highly heterogeneous, with patients showing diverse patterns of structural and functional disruption. A key challenge is capturing this variability in a way that supports individualised prognosis and treatment.

In our lab, we apply high-dimensional statistical and machine learning approaches to characterise how far an individual patient deviates from healthy control populations in “connectivity space”. This allows us to quantify patient-specific patterns of brain network disruption rather than relying on coarse group-level summaries.

We also use machine learning methods to identify neuroimaging and connectivity features that help explain and predict cognitive recovery after brain injury. While still an emerging area of work, these approaches extend our broader interest in personalised brain network modelling and outcome prediction across neurological disorders.