Multisite data harmonization

Diffusion imaging provides insights into the brain network architecture that can be studied developmentally. Multisite diffusion imaging data are increasingly becoming available, which facilitates development of methodologies to harness big data to significantly increase sample size and statistical power. This also enables life-span studies that span high age ranges from infancy to elder age. However, pooling data acquired from different is challenging due to the variation in acquisition parameters. Data harmonization schemes that facilitate a combination are essential. We are creating various schemes to correct for inter-site differences while preserving biologically relevant variation in the data.

Collaborators:
Publications
  1. J.-P Fortin, D. Parker, T. Watanabe, M. Elliott, K. Ruparel, R. C. Gur, R. E. Gur, R. T. Schultz, R. T. Shinohara, R. Verma, Statistical harmonization of multi-site diffusion tensor imaging data with ComBat, accepted ISMRM 2017
  2. T. Watanabe, B. Tunç, D. Parker, J.-P. Fortin, M. Elliott, K. Ruparel, R. C. Gur, R. E. Gur, R. T. Schultz, R. T. Shinohara, R.Verma, Normalization of inter-site Structural Connectivity Data for Regression analysis, accepted ISMRM 2017