Recognition of Infected Erythrocytes by Inclusion Tree Representation and Parasitemia Estimation in Blood Smear Images

Doke Pranoti R., Doke Pooja R
2019 International Journal of Scientific Research in Science and Technology  
Blood cells are composed of erythrocytes (red blood cells, RBCs), leukocytes (white blood cells, WBCs) and thrombocytes (platelets). Both WBC and RBC have fixed count in our body. If their count is less than the ideal count then it is an indication that our body is not healthy. Hence blood count helps in detecting many diseases in early stage.</p> <p>According to World Health Organization about 3.2 billion people are at risk of malaria[2]. But, malaria is preventable and curable, if the patient
more » ... is correctly diagnosed in early stage.</p> <p>The proposed approach to diagnose malaria mainly consists of following steps:</p> <ol> <li>Preprocessing, Histogram and Segmentation</li> <li>Inclusion-Tree representation</li> <li>Splitting of clumped erythrocytes</li> <li>Counting and labeling</li> <li>Cell stage identification</li> <li>Feature extraction and classification</li> </ol> <p>The algorithm is used to count malaria infected RBC in blood smear. A clump splitting method is used for precise RBC counting. Cell stage identification is performed in this approach by calculating Equivalent Circular Diameter. Quantification method improves overall performance in the determination of stages of infection such as ring, trophozoite and Schizont. Percentage of Parasitemia is calculate.
doi:10.32628/ijsrst196149 fatcat:fsjou4combe4digu5fopxra2ui