Optimizing White Blood Cell Nuclei Segmentation: A Proficient Method for Improved Diagnostic Accuracy

Clinical evaluations involving blood testing are the most crucial and often requested tests. If a blood sample seems odd, it must be examined manually under a microscope. This is a tedious, random, and time-consuming manual process. Automating the categorization of microscopic blood samples might be advantageous in order to speed up and improve the accuracy of pathologists' job. Segmentation is the first and most important step in the automated classification of blood cells. In this work, we present a practical technique for automatically dividing blood cell nuclei. Grayscale contrast enhancement and filtering are the foundation of the technique. There is now a minimum section size in place to get rid of fake items. 365 images of blood are used to evaluate the procedure. A quantitative assessment of the test set revealed a 79.7% segmentation performance. This performance is outstanding when compared to other published methods that have been applied to the same dataset. We compare the individual functions of the five typical kinds of white blood cells. Eosinophils have the lowest segmentation accuracy (69.3%), whereas monocytes have the best (86.3%). For comparison and replication, the blood image dataset and MATLAB source code are accessible.