Meeting Banner
Abstract #2561

Segmentation of Thalamus by Clustering of Resting-State fMRI

Yi-Ping Chao1,2, Chun-Yi Lo3, Ching-Po Lin4

1School of Applied Information Sciences, Chung Shan Medical University, Taichung, Taiwan; 2Department of Medical Imaging, Chung Shan Medical University Hospital, Taichung, Taiwan; 3Department of Medical Imaging & Radiological Sciences, National Yang Ming University, Taipei, Taiwan; 4Institute of Neuroscience, National Yang Ming University, Taipei, Taiwan


In this study, we hypothesized that voxels belonging to the same partition of thalamus should be provided with the similar temporal cross-correlation values to other voxels. Therefore, we transferred the cross-correlation matrix derived from resting-state fMRI analysis to a meta-data for the data mining software to separate this data based on the k-median clusting algorithm. Finally, we presented the segmentations of thalamus according to the clustered results. From our results, resting-state fMRI could provide not only the functional connectivity network between cortical and subcortical brain regions, but also local characteristic within thalamus.