报告人：Gang Li 副教授（UNC）
报告题目：Learning-based Quantification of Baby Brain Development
报告摘要: The increasing availability of infant brain MRI data, such as the data collected from the Baby Connectome Project (BCP), affords unprecedented opportunities for precise charting of dynamic early brain developmental trajectories in understanding normative and aberrant brain growth. However, most existing neuroimaging analysis tools, which are mainly developed for adult brains, are not suitable for infant brains, due to extremely low tissue contrast and regionally-heterogeneous dynamic changes of imaging appearance, brain size, shape and folding in infant brains. In this talk, I will introduce a set of our pioneered machine learning based neuroimaging computational methods for quantitatively characterizing baby brain development, including tissue segmentation, cortical topological correction, surface parcellation, and missing data estimation and prediction. I will also show neuroscience applications of these methods in advancing our understanding of the baby brains.
Dr. Li is the assistant professor in Biomedical Research Imaging Centre, School of Medicine, University of North Carolina at Chapel Hill (UNC). He received his PhD from Northwestern Polytechnical University in 2010. He was research fellow in Weill Medical College of Cornell University and Harvard Medical School in 2007-2008 and 2005-2007 respectively. His research focuses on the development of novel computational techniques for quantitative analysis of biomedical images. He has created a variety of novel methods and tools for advanced cortical surface-based neuroimaging analysis, including cortical surface reconstruction, parcellation, modeling, measurement, mapping, sulcal-gyral landmark curve extraction and atlas building, with applications to studying brain development, aging and disorders. Recently, he has been working on 4D cortical surface based neuroimaging analysis tools and atlases for accurate characterization of the dynamic, nonlinear and regionally-heterogeneous early brain development in typically-developing infants and infants at high-risk for neurodevelopmental disorders, such as schizophrenia and autism. His computational tools and discoveries on infant brain development, published in The Journal of Neuroscience in 2014, has been highlighted in NIMH’s 2015-2020 Strategic Plan.