Search Papers | Poster Sessions | All Posters
Poster A125 in Poster Session A - Tuesday, August 6, 2024, 4:15 – 6:15 pm, Johnson Ice Rink
A Multi-Modal Neuroimaging Study on the Prediction of Alcohol Sipping Patterns in Children: Results from the ABCD Study
Ana Ferariu1, Hansoo Chang1, Alexei Taylor1, Fengqing Zhang1; 1Department of Psychological and Brain Sciences, Drexel University
As individuals transition from childhood to adolescence, alcohol sipping and drug initiation increases. Early alcohol exposure can lead to risky alcohol consumption and alcohol dependence later in life. We previously used data from the Adolescent Brain and Cognitive Development Study (ABCD) to detect latent alcohol sipping trajectories over time. In the current study we examined brain imaging data measured at baseline as a potential biomarker in predicting alcohol sipping patterns. We used several popular machine learning methods on structural (sMRI) and resting-state functional magnetic resonance imaging data (rs-fMRI) separately and then combined to detect important features that can predict alcohol sipping in children aged between 9 and 14. Ridge regression showed the best performance and results show that the latent alcohol sipping groups can be better predicted by rs-fMRI data than by sMRI data at baseline.
Keywords: multi-modal brain imaging alcohol sipping machine learning ; longitudinal analysis