ABSTRACT
Autoimmune diseases (ADs) are often diagnosed via indirect immunofluorescence (IIF) with human epithelial type-2 (HEp-2) cells. Computer aided diagnosis systems and automatic classification of HEp-2 cells can improve the diagnostic process in terms of lower cost, faster response, and better repeatability. In this chapter, we propose an adaptive distributed dictionary learning (ADDL) method where the dictionary learning problem is reformulated as a distributed learning task. Using this approach, we develop an automatic and robust method that effectively handles the complexity of the problem in terms of memory and computational cost. To improve the classification accuracy, we combine SURF (speeded-up robust features) and SIFT (scale-invariant feature transform) in a complementary fashion. The performance of our method is evaluated on two data sets and is shown to outperform state-of-the-art techniques in both classification accuracy and computational costs.
