The medical industry has made significant strides thanks to the use of many autonomous systems to identify various ailments in this day, surrounded by numerous technology. A crucial medical practice is the visual evaluation and counting of white blood cells in microscopic peripheral blood smears. It may offer helpful details about the patient’s health, such as the identification of Acute Lymphatic Leukemia or other serious illnesses.
In this study, a framework for recognizing acute lymphoblastic leukemia from a white blood cell's microscopic picture is proposed. Microscopic images must first undergo a thorough pre-processing step in order to be classified. In this study, a collection of textural, geometrical, and statistical features are retrieved from the segmented region following the segmentation of WBCs from blood smear pictures and morphological procedures.
In order to compare these algorithms in terms of various performance measures, four distinct machine learning techniques—namely, Random Forest (RF), Support Vector Machine (SVM), Naive Bayes classifier (NB), and K nearest neighbor (KNN) are also deployed. After careful comparison, it can be seen that the SVM is effective at classifying and identifying the acute lymphoblastic cell that causes leukemia malignancy.
A single classifier is nearly useless given the variety of blood smear pictures. As a result, we thought about using an EMC-SVM to classify leukocytes. The suggested method properly separates WBCs from blood smear images, according to experimental findings, and correctly classifies each segmented cell into its relevant category, which includes neutrophil, eosinophil, basophil, lymphocyte, and monocyte.