Join us every Friday at 12:00 p.m. EDT for BME Breaks, Columbia University's weekly webinar series hosted by the Department of Biomedical Engineering. Don't miss the opportunity each week to hear from global leaders in Biomedical Engineering research!

ABOUT THE MAY 28 WEBINAR

Developing Connectome-Based Predictive Models of Behavior

Technical advances and large-scale data sets have allowed for the development of models capable of predicting individual differences in traits and behavior using brain connectivity measures derived from neuroimaging data. Here, I will discuss our efforts to improve predictions of phenotypic information from functional connectivity data. These include novel methods to include combining connectomes from diverse sources, creating latent behavior phenotypes from multiple behavioral measures, data imputation to resolve missing data problems, and methods to transform connectomes of one size and type into another. I will show have these approaches can be used to a wide-range of predict phenotypic data that is relevant for mental health disorders and cognitive neuroscience.

ABOUT THE SPEAKER

Dustin Scheinost, Ph.D., Assistant Professor of Radiology & Biomedical Imaging, Biomedical Engineering, Statistics & Data Science, Yale Child Study Center

Dustin Scheinost, Ph.D., is an Assistant Professor of Radiology & Biomedical Imaging, Biomedical Engineering, Statistics & Data Science at the Yale Child Study Center. The Multi-modal Imaging, Neuroinformatics, & Data Science (MINDS) Lab’s research is three-fold. First, using state-of-the-art research for connectomics, we aim to develop novel statistical and machine learning methods to meet challenges arising with the “big” neuroscience data. Second, the MINDS lab helps develop BioImage Suite Web (BISWeb; https://bioimagesuiteweb.github.io/webapp/), an image analysis webapp. Third, we are at the cutting-edge of neuroimaging in fetuses and infants and are a member of Fetal, Infant, Toddler Neuroimaging Group (FIT’NG).