Images have become critical components for a detailed understanding of the structure and function of cells and proteins. For example, confocal microscopy images are used to obtain a better understanding of several critical functions of the central nervous system (CNS), such as mechanisms behind the loss and recovery of vision following retinal detachment. Image processing and information discovery tools increasingly play a significant role in better harvesting these vast amounts of data, most of which is currently analyzed manually and qualitatively. The main contribution of this work is the development of quantitative analysis and modeling of bio-molecular images. We focus primarily on retinal images for characterizing patterns of cellular/ subcellular protein distribution and changes in such patterns. Specifically, we develop an automated nucleus detection method which provides reliable and accurate results for counting cell nuclei. Quantitative measurements to characterize the structural distortion of the retina, including layer thickness, local cell density, and distortion indices, are proposed. We model the retinal detachment process using a Bayesian network in order to understand the protein-protein interactions within the retinal cells as well as interactions between cells. Quantitative analysis and statistical modeling provide opportunities to test therapeutic agents that may reduce the damaging effects of detachment or improve the outcome of reattachment surgery.