In this paper we propose a medical image integrity verification system that not only allows detecting and localizing one alteration, but also provides an approximation of this latter. For that purpose, we suggest the embedding of a signature derived from Geometric moments of image pixel blocks. Image integrity verification is conducted by comparing this embedded signature to the recomputed one. This signature helps to approximate modifications by determining the parameters of the nearest generalized 2D Gaussian. Experimental results with local image modification illustrate the overall performances of our method.