Department of Psychiatry
University of Pennsylvania School of Medicine
Increasingly, large-scale studies of brain development incorporate complex imaging data alongside diverse clinical and cognitive measures. However, it remains a challenge to integrate these high-dimensional data, and generate measures tailored to the individual. Here I will focus on recent work illustrates how data integration and personalization can inform our understanding of normal and abnormal brain development. Specifically, I will demonstrate how machine learning techniques can help identify links between dimensions of psychopathology in youth and abnormalities within functional brain networks. Furthermore, I will present unpublished work that uses a combination of functional imaging data and machine learning techniques to identify personalized networks in the developing brain. Together, such approaches may evolve to help produce biomarkers of risk and resilience in youth.