Detection of Anaemia using Image Processing Techniques from microscopy blood smear images

Abstract

The human blood includes Red blood Corpuscles (RBC), White Blood Corpuscles (WBC), platelets and plasma. The status of one’s health is determined by a complete blood count, therefore segmentation and identification of blood cells are critical. A Complete Blood Count (CBC) is a test that counts all of the cells in the body to assess a person’s health. The RBC and WBC count are vital in diagnosing disorders such as anaemia, leukaemia, tissue damage, and so forth. This paper focuses mainly on RBC counting and the detection of abnormality, anaemia based on the count of RBCs from a peripheral blood smear using digital image processing techniques. Anemia is an indicator, and the most important one at that, for many other diseases. Therefore, basic screening of anemia is very important, especially in regions prone to poverty. Malnutrition due to poverty is the major cause for anemia. The paper presents an algorithm to automatically count the RBCs present in the blood of a person. The count of RBC, in 1 microlitre if blood is considered and it is observed how the count varies in normal blood smears and the anemic blood smears. In remote places, where a lot of people are to be screened, using cell counters and hemocytometer is not feasible. A faster method of counting is one of major demand. Therefore, to reduce the computation time, an algorithm in digital image processing is developed to compute the number of red blood cells. Although anemia is a vast subject and there are various different characteristics to consider, this is a humble approach to automate the counting of the RBCs which would be useful for future research purposes.