I am interested in using mathematics and statistics to quantify the capability of imaging systems to describe biological processes in an accurate and precise manner, and then using these metrics to optimize the systems in question.
I have focused my recent efforts on studying the maximum precision a combined balanced steady-state free precession (bSSFP) and spoiled gradient-recalled echo (SPGR) modality could provide for estimates of myelin water fraction within a voxel when a two-pool model is assumed. Entirely a computational endeavor (so far), the process entails calculating the Cramer-Rao lower bound of the variance of the estimated signal fraction through the Jacobian sensitivity matrix. I am also investigating the possibility of derivative models which provide biased estimates with lower mean-squared errors. By constraining certain model parameters, the covariance in fitted parameters can be decreased dramatically, resulting in an overall more precise fitting at the cost of a bit of accuracy.