Shantanu Majumdar1, David C. Zhu2,3, Guy Raguin3,4, Satish S. Udpa1
1Electrical and Computer Engineering, Michigan State University, East Lansing, MI, USA; 2Psychology, Michigan State University, East Lansing, MI, USA; 3Radiology, Michigan State University, East Lansing, MI, USA; 4Mechanical Engineering, Michigan State University, East Lansing, MI, USA
In Diffusion Tensor Imaging, an assumption of isotropic diffusivity in the direction transverse to fiber orientation can be applied to the diffusion model to create an axisymmetric DTI model. An optimization procedure for the selection of diffusion encoding gradients by using a priori information of the anatomical structure has been presented for the axisymmetric DTI model. The optimization applies a priori information to a D-optimality based method to compute a set of gradient directions that reduce the uncertainty in the estimate of ADTI model parameters. In this work, a region of the cervical spinal cord has been imaged with an optimized gradient scheme. The data was used to estimate the covariance matrix of the estimation error for the model parameters. It was demonstrated that the optimized scheme provides a lower overall variance in the estimates than that offered using a standard scheme.