NEIGHBORHOOD GRAPHS AND IMAGE-PROCESSING - APPLICATION TO IMAGES OF IMMUNOHISTOCHEMICAL STAINING

Devising an image analyzer dedicated to the automatic quantification o f immunohistochemical staining for clinical oncology implies developin g a method for the delimitation of tumoral cell nests, setting aside t umoral stroma, while accounting for the topology of the staining. The representation of images by neighborhood graphs can bring an answer to both requirements. In this paper, a methodological approach is presen ted. It consists in a preliminary study dealing with nuclear immunosta ining images of breast cancer. Segmentation of the graph structure all ows to separate clusters of cancer cells and the analysis of this stru cture can account for the focal or diffuse aspect of the staining with in the tumor.