Abstract: Our long-term goal is to improve telepathology images for clinical use. Over the years we have carried out a series of studies addressing a variety of key issues associated with the interpretation of virtual pathology slides. These studies will be reviewed in this lecture. The initial studies were designed to demonstrate that diagnostic accuracy was equivalent between virtual and traditional (light microscopy) viewing. Then we sought to understand how pathologists view virtual slides in the digital reading environment using eye-position recording techniques. More recently we have investigated two aspects of image quality. In the first study we compared diagnostic accuracy using a color-calibrated, color-managed display compared to the same display without any special calibration (“out-of-the-box”). Surprisingly the sophisticated calibration technique had little to no impact on accuracy but did shorten viewing times somewhat. In another series of studies we have been investigating how much we can compress virtual slides before it impacts diagnostic accuracy. In these studies we also aim to demonstrate the utility of a visual discrimination model for predicting observer performance. In the most recent study, observer performance (Az) was nearly constant up to a compression ratio of 32:1, then decreased significantly for 64:1 and 128:1 compression. Virtual pathology may be compressible to relatively high levels before impacting diagnostic accuracy and the VDM accurately predicts human performance. An eye-position study followed demonstrating significant changes in scanning properties as a function of compression.