AUTOMATED LEUKEMIA DETECTION USING HYBRID DEEP LEARNING AND MEDICAL IMAGE ANALYSIS

Abstract

Leukemia is a life-threatening cancer of the blood and bone marrow, where early detection plays a crucial role in improving patient outcomes. Traditional methods of examining blood smears under a microscope are time-consuming and prone to human error. This study explores the use of deep learning techniques, specifically convolutional neural networks (CNNs), to automatically analyze blood smear images and detect signs of leukemia. The proposed system is trained on labeled datasets to identify abnormal white blood cells with high accuracy. Experimental results demonstrate that deep learning models can significantly enhance the speed and reliability of leukemia diagnosis, offering an effective decision-support tool for hematologists and reducing diagnostic delays. . A CNN architecture, such as ResNet or VGG16 with transfer learning, will be trained and validated to classify blood smear images as healthy or leukemic. By training models on large annotated datasets, the system can accurately classify and highlight abnormal leukocytes, significantly reducing diagnostic time and increasing consistency. Experimental results demonstrate that deep learning approaches can achieve high sensitivity and specificity, indicating their potential to assist hematologists in early and reliable leukemia diagnosis.