32nd IEEE Conference on Signal Processing and Communications Applications, SIU 2024, Mersin, Türkiye, 15 - 18 Mayıs 2024
In this study, we propose a new model for visualizing brain networks of brain using functional Magnetic Resonance Imaging (fMRI). In this model, we estimate the probability density functions of the regions by considering the voxel time series in the anatomical regions as random variables. Using these probability density functions, we estimate the Kullback-Leibler divergences between anatomical region pairs to assess the connection strength between regions. Using KL divergence values as the connection weights, we construct dynamic brain networks that illustrate information exchange among brain regions. The nodes of this network correspond to anatomical region indices, while the edge weights are defined as KL divergences. Finally, the brain networks developed for early and advanced-stage Alzheimer's disease patients and healthy individuals were visualized with BrainNet Viewer software.