Epi-fluorescent image modeling for viral infection analysis

Quantitative analysis in fluorescent microscopy is increasingly being employed to understand complex biological processes. However, fluorescent microscopy is beset with several inherent distortions that require pre-processing of fluorescent images prior to any meaningful quantitative analysis. None of the previous techniques for fluorescent microscopy comprehensively address all the distortions and their corrections. In this paper, we present a fluorescent image model that serves as a framework for a fully automated retrospective denosing process. Immunofluorescent intensity signals (IIS) derived from denoised images provide accurate measurement of relative protein concentration and distribution not possible with IIS obtained from noisy images.