Detection of Blood Cancer Cells using SVM Algorithm

Information Technology

Cancer is a fatal illness. A type of cancer known as blood cancer targets the lymphatic, bone marrow, or blood systems. Hematologists do a blood test to diagnose it, during which certain blood cell types are examined. Currently, there are numerous varieties of blood cancer cells like Leukaemia, Lymphoma, Myeloma, etc. Leukemia is a type of blood cancer that develops when White Blood Cells (WBCs) are abnormal or immature. We only looked at acute myelogenous leukaemia, which is one of the blood cancer types that falls under the category of acute leukaemia and most frequently affects adults. Leukemia must be automatically recognized since when doctors see abnormal blood cells under a microscope using CBC (Complete Blood count), it requires a lot of manual work and time so it may harmful to the patient. In recent years, Support Vector Machines (SVM) have gained popularity in the field of medical image analysis due to their ability to accurately classify and segment data. In this study, we propose an approach to detect blood cancer cells using SVM. The proposed method involves preprocessing of the blood smear images to extract features, such as shape, texture, and color. The extracted features are then used as input to train the SVM classifier, which classifies the cells as normal or abnormal. The performance of the SVM classifier is evaluated using metrics such as sensitivity, specificity, accuracy, and F1-score. The proposed approach has the potential to aid in the early detection and diagnosis of blood cancer, which could improve the overall survival rate of patients.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.