Classifying Child Maltreatment by Brain Imaging Data

Thesis (Master's)--University of Washington, 2016-08

Childhood maltreatment is a condition that leads to the development of behavioral issues and affect brain structure and functionality. Recent neuroimaging studies provide connection between deficits in brain volume, gray and white matter of several cortical regions, such as dorsolateral and ventromedial prefrontal cortex and also hippocampus and amygdala. Machine learning has become the major technique in brain imaging and also computational neuroscience, since they are very seminal for training great amount of neural data of increasing measurement precision and acquiring signals from very noisy data. Machine learning techniques provide the tools to characterize brain states and distinguish them from non-informative brain signals. In this thesis, I use machine learning to classify childhood maltreatment using MRI scans of adolescents and children. The main goal of this thesis project is to find methods to classify childhood maltreatment and find predictive brain regions that contribute to it. I also look at other features such as age and gender, to see how predictable they are based on the brain imaging data.