Implementation of an image analysis system for morphological description of skin mast cells in urticaria pigmentosa

摘要

The aim of the study was to develop a new method to describe the morphology and localization of mast cells in urticaria pigmentosa and healthy skin. We employed Astra blue stain to visualize tissue mast cells with subsequent unique identification of those cells within the whole picture. The procedure consisted of the following steps: 1. Astra blue staining of the mast cells allowed us to threshold the image and as a result mast cell objects were obtained marked in black on a white background. 2. Geometrical features such as area, edge and maximal diameter of the mast cell objects were calculated, 3. The dermoepidermal junction was considered as a texture and for this texture a set of statistical features was calculated; image segmentation resulted in an image region with a clearly separated dermoepidermal junction. 4. The Euclidian distance was calculated from the mast cell and the dermoepidermal junction.Astra blue stain allowed the differentiation between mast cells and the surrounding tissue elements. The employed methods of image analysis allowed quick, automated characterization of skin mast cells. In conclusion, it should be pointed out that the employed image analysis techniques may be of value in describing localization and morphological parameters of mast cells. Refinement of the method may be useful to image other cell types as well.