Howard Kushner, Ph.D.
Dr. Kushner’s main research interest is the statistical analysis of imaging data. Three research projects are:
1. The Maximum Likelihood (ML) Method in Diffusion Tensor Imaging. The aim is to justify the ML method (presently used uncritically) in DTI imaging analyses.
2. Mean Diffusivity in an Optimal and Uniform Diffusion Tensor Imaging (DTI) Design. This research shows that the standard uniform arrangement of gradient vectors (a “uniform design”) is not optimal. It also determines the efficiency of a uniform DTI design, and the optimal ratio of the number of DTI observations with and without gradients in an optimal design and in a uniform design.
3. The Image Intra-class Correlation Coefficient (I2C2) for Multivariate Ratings. The goal of this research is a statistical analysis of a reliability measure (I2C2) for the multivariate ratings produced by different MRI scans.
Dr. Kushner’s previous research interest was Optimal Repeated Measurements Designs. (Project 2 above can be considered to be a project in “DTI Experimental Designs”.)
A fourth project is:
4. Efficiency of Repeated Measurement Designs for Correlated Observations. This research concerns the problems that result when the correlated observations in an experiment are not treated properly.
KUSHNER, HB. (2003) Allocation rules for adaptive repeated measurements designs. The Journal of Statistical Planning and Inference, 113: 293-313.
KUSHNER HB. (1999) H-symmetric optimal repeated measurements designs. The Journal of Statistical Planning and Inference, 76:235-261.
KUSHNER HB. (1998) Optimal repeated measurements designs for uncorrelated observations. The Journal of the American Statistical Association, 93:1176-1187.
KUSHNER HB. (1997) Optimal repeated measurements designs: The linear optimality equations. The Annals of Statistics, 25:2328-2344.
KUSHNER HB. (1997) Optimality and efficiency of two treatment repeated measurements designs. Biometrika, 84:455-468.