Image similarity can be deened in a number of diierent semantic contexts. At the lowest common denominator, images may be classiied as similar according to geometric properties, such as color and shape distributions. At the mid-level, a deeper image similarity may be deened according to semantic properties, such as scene content or description. We propose an even higher level of image similarity, in which domain knowledge is used to reason about semantic properties, and similarity is based on the results of reasoning. At this level, images with only slightly diierent (or similar) semantic descriptions may be classiied as radically diierent (or similar), based upon the execution of the domain knowledge. For demonstration, we show experiments performed on a small database of 300 images of the retina, classiied according to fourteen diagnoses.