Detection of Abnormalities in Blood Sample Using WBC Differential Count

Abstract

The emergence of smart healthcare has immensely revolutionized the traditional approaches of disease detection and diagnosis. Smart health system is characterized by innovative methods that offer better diagnostic instruments and enhanced treatment for patients with gadgets improving lifestyles. The doctors can analyze the blood reports of patients to check the normal functioning of body parameters as well as detect abnormalities present in cell population. The significance of our proposed model is to analyze blood samples and make the patient aware of potential risks they might possess toward hematological diseases. In the bloodstream, the density of white blood cells (WBC) signifies whether the immune system is vulnerable toward infections or ailments. Sudden drastic variation in WBC count corresponding to threshold value often signifies the presence of antigens in the body. Change in a particular type of WBC usually correlates to the existence of specific type of antigen. Deriving differential count of WBCs in the bloodstream can provide powerful mechanism to introspect health conditions. Apparently, differential blood count is an expensive and prolonged task that requires in-depth medical analysis and scrutiny. Our novel approach involves designing an automated system to classify blood cell types using image processing techniques and deep learning, followed by the detection of cell deformities.