Abstract: In this work we propose an approach to select the classification method and features, based on the state-of-the-art, with best performance for diagnostic support through peripheral blood smear images of red blood cells. In our case we have used samples of patients with sickle-cell disease which can be generalized for other study cases. To trust the behaviour of the proposed system, we have also the overlapped cells into a single cell, one can get the exact blood count and all the relevant blood statistics like RBC and WBC etc. analyzed the interpretability. We have pre-processed and segmented microscopic images, to ensure high feature quality. We have applied the methods used in the literature to extract the features from blood cells and the Deep learning methods to classify their morphology. Finally, comparing the best performing classification methods with the state-of-the-art, we obtained better results even with interpretable model classifiers.
Key Word: Red blood cells, RBC and WBC, Morphology