Machine learning based Diagnosis and Classification Of Sickle Cell Anemia in Human RBC

Anemia is a disease which is caused by the deficiency of red blood cells. The shape of red blood cell changes to sickle or crescent shape in sickle cell anemia disease. The manual inspection of microscopic images is very difficult and time-consuming process. In this research image processing and machine learning techniques is used to automate the process of detection of sickle cells in microscopic images then classify the RBC into three shapes: circular, elongated (sickle cell) and other shape. The microscopic image is pre-processed and Otsu thresholding technique is used for segmentation. Then, Watershed segmentation is applied to separate the overlapped cells. The geometrical, statistical and textural features are extracted from the images. The machine learning classifier random forest, logistic regression naïve baye sand support vector machine is used. This research describes the comparison among these algorithms.