Human epithelial type-2 cell categorization using hybrid descriptor with binary tree

Indirect immunofluorescence (IIF) method with human epithelial type-2 (HEp-2) cell as substrates is recommanded in anti nuclear antibody (ANA) test. These IIF slides are observed under microscope by pathologists to prepare the report. So, the ANA test is subjective and requires systematic automation. This paper proposes a novel algorithm for HEp-2 cellcategorization, which can be implanted in the ANA test automation system. A hybrid descriptor which represents the textural and morphological characters of the objects of interest was used along with Binary tree. The hybrid descriptors were generated by decomposing the image into binary images by using lower and upper thresholds. The performance of the proposed algorithm was evaluated on the ICPR 2016 IIF HEp-2 cell image dataset, the results concluded that Hybrid Descriptor with Binary Tree approach achieved the best performance with “97.5%” mean class accuracy.