Morphology Based Automatic Disease Analysis through Evaluation of Red Blood Cells

Cell morphology has been an active area in the field of bio-medical research. When applied for blood microscopic images, one can study blood cell characteristics and detect abnormalities. In this paper, we introduce an automatic, cost effective and accurate way of red blood cell analysis and evaluation through Blob detection, Morphology operations and Hough circle transformation techniques for identification of four common types of anemia. Our research has filled the gaps in the existing literature by developing an integrated system to Count RBC, Diagnose Elliptocytes, Microcytic, Macrocyte and Spherocytes Anemia, Detect abnormalities and Separate overlapped cells, automatically, accurately and efficiently. The result shows an insight in the manually processed results with 99.545% accuracy of RBC count. Each sub method is closely running in the range 91%-97% of accuracy. The achievements are highlighted as efficiency through automation, cost effective, elimination of human error and easy to manipulate.