Copyright © 2005 Hindawi Publishing Corporation. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Tissue microarrays (TMAs) make possible the screening of hundreds of different tumour samples for the expression of a specific protein. Automatic features extraction procedures lead to a series of covariates corresponding to the averaged stained scores. In this article, we model the random geometry of TMA cores using voronoi tesselations. This formalism enables the computation of indices of spatial correlation of stained scores using both classical and novel approaches. The potential of these spatial statistics to correctly discriminate between diseased and non-diseased cases is evaluated through the analysis of a TMA containing samples of breast carcinoma data. The results indicate a significant improvement in the breast cancer prognosis.