Anemia is a disease caused due to the lack of healthy Red blood cells. Red blood cells are mainly responsible for carrying oxygen to different parts of the body. It releases generated red blood cells into the bloodstream after the proper maturation. Anemia is primarily classified as chronic or acute. Chronic anemia continues to exist for a long time. Acute anemia occurs rapidly. It causes sickle cell anemia due to red blood cells that are in shape of a sickle or a crescent moon. The irregular shape makes it get stuck in the bloodstream which leads to a reduced flow of oxygen. The primary aim of this work is to overcome the problems involved in the manual process of detection of anemia. Manual processes require proper pipetting skills and hence is imprecise. In the proposed work, an efficient technique using a decision system is proposed that helps the pathologists to identify the disease accurately. The decision system is built by deriving the rules from the decision tree. The granulometric analysis is carried out to separate the Red blood cells from the other components of the blood. Modified Watershed transform algorithm is used to isolate cluttered cells. The work defines a systematic approach to categorize the red blood cells with central pallor and without central pallor. It puts forward the comparison of the work with state-of-the-art methods. It shows the role of the accurate segregation of Red blood cells in getting the blood cell count. The precise separation of the cluttered cells improved the accuracy of the system. Results show that the work achieved an overall accuracy of 98.5% in the identification of Sickle cell anemia and an efficiency of 97.6% in the separation of RBCs.
|Number of pages||14|
|Journal||Journal of Engineering Science and Technology Review|
|Publication status||Published - 01-01-2019|
All Science Journal Classification (ASJC) codes