Classification of HEp-2 cells using fluorescent image analysis and data mining

Abstract

The cells that are considered in this application for an automated image analysis are Hep-2 cells which are used for the identification of antinuclear autoantibodies (ANA). Hep-2 cells allow for recognition of over 30 different nuclear and cytoplasmic patterns, which are given by upwards of 100 different autoantibodies. The identification of the patterns has recently been done manually by a human inspecting the slides with a microscope. In this paper we present results on image analysis, feature extraction, and classification. Starting from a knowledge acquisition process with a human operator, we developed an image analysis and feature extraction algorithm. A data set containing 162 features for each entry was set up and given to a data mining algorithm to find out the relevant features among this large feature set and to construct the classification knowledge. The classifier was evaluated by cross validation. The results show the feasibility of an automated inspection system.