Abstract #3040

Multimodal brain MRI at birth predicts neurodevelopmental outcome at 2 years of age

Minhui Ouyang¹, Qinmu Peng¹, Lina Chalak², Nancy Rollins³, and Hao Huang^1,4

¹Department of Radiology, The Children's Hospital of Philadelphia, Philadelphia, PA, United States, ²Department of Pediatrics, University of Texas Southwestern Medical Center, Dallas, TX, United States, ³Department of Radiology, University of Texas Southwestern Medical Center, Dallas, TX, United States, ⁴Department of Radiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, United States

Structural and functional maturation level of brain at birth can be quantified with multimodal MRI, including DTI and resting-state fMRI (rs-fMRI). We hypothesized that features extracted from multimodal brain MRI at birth could better predict the neurodevelopmental outcomes at 2 years of age compared to features extracted from only DTI or rs-fMRI. With combined features of white matter fractional anisotropy and cortical functional connectivity strength from neonatal DTI and rs-fMRI respectively, higher accuracies were achieved using machine learning models to predict Bayley scores at 2 years of age. Heterogeneous feature-contribution patterns were observed across cortical and white matter regions.

This abstract and the presentation materials are available to members only; a login is required.

Join Here