This paper presents an image processing framework for assessing molecular activity changes from fluorescent data in time-dependent geometries. The aim of our work is to provide the necessary tools in order to facilitate molecular imaging studies of small animals, where the basic problem is the time dependent geometry due to different measurement sessions or animal movement. Synthetic, fluorescence imaging data of molecular activity inside a mouse are produced numerically by moving the location of the fluorophore distribution and then randomly and non-linearly transforming the image. A method for aligning the temporal data and improving the accuracy of monitoring fluorophore distribution is presented.
Kostas Marias, Stelios C. Orphanoudakis, Jorge Rip