An automated system for platelet segmentation using histogram-based thresholding

Platelet count is a crucial diagnostic metric for identifying several diseases. In most of the laboratories, Leishman stained blood slides are used for counting platelets. However, the problem arises when huge numbers of blood samples are to be tested by the laboratory technicians making the entire process time-consuming and prone to human errors. In this paper, we propose an automated system for counting platelets, which eliminates the need of expert lab technicians and reduces the cost and time of the test. We have used histogram based thresholding and morphological operations to extract platelets from the microscopic image of blood cells. After the platelets are extracted, our automated system uses a mechanism analogous to the one that is used in laboratories to count the number of platelets present per microliter of blood for each patient. Through a comparative study of manual platelet count and the platelet count given by this automated system, we found that the automated system present here gives near accurate result and is highly efficient.