Meeting Banner
Abstract #3578

Outlier Detection Based on the Neural Network for Tensor Estimation

Zhenyu Zhou1, 2, Yijun Liu3, Guang Cao1, Karen M. von Deneen3, Dongrong Xu2

1Global Applied Science Laboratory, GE Healthcare, Beijing, China; 2MRI Unit, Columbia University, New York, NY, United States; 3McKnight Brain Institute, University of Florida, Gainesville, FL, United States

Diffusion weighted imaging is always influenced by both thermal noise and spatially and/or temporally varying artifacts such as subject motion and cardiac pulsation. Motion artifacts are particularly prevalent, especially when scanning an uncooperative population with several disorders. In this study, we proposed a classifier work frame which can classify DWIs as normal images or motion artifacts. It achieves better performance in tensor estimation by automatic unvoxel-wise outlier rejection compared with manual and visual inspection, and previous voxel-wise outlier rejection methods. The proposed method could potentially remove the influence of unexpected motion artifacts in DWI acquisitions.