Identifying Subtypes of Acute Lymphoblastic Leukemia Using Blood Smear Images: A Hybrid Learning Approach

Leukemia is a type of blood cancer that affects a large number of people worldwide. Detecting and classifying leukemia is crucial in determining the treatment plan for patients and improving their chances of survival. The design of a hybrid model comprising MobileNet as a feature extractor and Support vector machine algorithm to classify the leukemia cells into four classes: benign, pre-B, early pre-B, and pro-B. The confusion matrix calculates various performance metrics such as F1 score, accuracy, recall, and precision in this method. The experimental results show that the suggested model performs better than the existing state-of-the-art models for leukemia cell classification, obtaining a remarkable existing state-of-the-art model for leukemia cell classification, obtaining an accuracy in classification of 99.3%. The proposed approach can aid in the early detection of leukemia, leading to better treatment outcomes and improved patient survival rates.