Abstract #4673

Artificial Neural Network Analysis of Differences in Fiber Tracks Between Term and Preterm Children

Lutfi Tugan Muftuler¹, Ke Nie¹, Orhan Nalcioglu¹, Christine E. McLaren², Min Ying Su¹

¹Center for Functional Onco-Imaging, University of California, Irvine, CA, USA; ²Department of Epidemiology, University of California, Irvine, CA, USA

VBM techniques can be applied to DTI parameter maps to investigate local differences in white matter structures between patient and control groups. However, tractography is essential to investigate tract morphology and brain connectivity. Here, we developed an Artificial Neural Network based analysis to select a set of features that can achieve the highest differentiation power between the two groups. Once these features are found, inferences about the morphological differences in fiber tracts can be made. Compared to traditional statistical analysis methods, ANN was found to have higher prediction rates in complex and non-linear relationships among a large number of variables.