Tissue section analysis: Feature selection and image processing

摘要

Three examples are given of the state-of-the-art of pattern recognition as regards the analysis of images of human and animal tissue: (1) the counting of cell nuclei of normal and abnormal rabbit kidney; (2) the detection of boundaries of endothelial cells of the living human cornea; and (3) the differentiation of normal and abnormal cells in the human liver. The methodology employed is that of cellular logic matched filtering using the Carnegie-Mellon SUPRPIC image processing system.