The remarkable development of digital photography has been possible thanks to joint advances in computer science and photography processing and retouching software. The scientific use of photography is now possible as long as its intrinsic limits are comprehended, particularly in the very specific time-limited structure of clinical trials. We list the two main limits inherent to this method: geometrical calibration and colorimetric calibration. These limitations are illustrated by numerous images used in our practice in Besançon during the last decade.