Automatic Segmentation of Red Blood Cells from Microscopic Blood Smear Images Using Image Processing Techniques

Human blood is a very effective parameter to detect, diagnose and rectify ailments of the human body. Complete blood count (CBC) is a method to clinically obtain a statistical measure of blood and its related parameters, i.e., red blood cells (RBCs), white blood cells (WBCs), platelets, hemoglobin concentration to name a few. This helps to determine the physical state of the subject. For further diagnosis, peripheral blood smear, a thin layer of blood smeared on a microscope slide and stained using various staining methods is examined for the morphology of the cells by the pathologists. However, manual inspection of smear images is tedious, time-consuming, and laboratorian-dependent. Although there are certain software-based approaches to tackle the problem, most of them are not robust for all staining methods. Thus, the need is to create an automated algorithm that will work for different staining types, thereby alleviating both the aforementioned drawbacks. This work aims to create an automatic method of segmenting and counting RBCs from blood smear images using image processing techniques to help diagnose RBC-related disorders. In the proposed method, the images are first preprocessed, i.e., standardized to a uniform color and illumination profile using contrast enhancement, adaptive histogram equalization followed by Reinhard stain normalization algorithms. WBCs and platelets are extracted in HSI color space and subtracted from the original image to retain only RBCs. Thereafter using morphological operations and active contour segmentation algorithms, a count of total RBCs were obtained even for overlapped cells in the microscopic blood smear image. The proposed method achieved counting accuracy of 89.6% for 150 images.