Scoring Histological Sections Through Immunohistochemistry

Immunohistochemical staining of biological samples provides a rapid protocol for visualizing various complexes and organelles. However, this method of sample staining is seldom used for direct quantitative analysis due to variations in sample fixations, ambiguities introduced by color composition, and the limited dynamic range of imaging instruments. We demonstrate that, through the decomposition of color signals, staining can be scored on a cell-by-cell basis. We applied our method to fibroblasts grown from histologically normal breast tissue biopsies obtained from two populations. Nuclear regions are initially segmented by a cascade of filters and geometric constraints. Subsequently, the strength of staining is quantified by a color decomposition model that is optimized by graph cut algorithm. By including color decomposition in the process, nuclear segmentation can be validated (corrected) for subsequent geometric representation. Finally, signal complexes are associated with each nuclear region following region-based Voronoi tessellation.