Leukemia, a type of blood cancer, is characterized by the abnormal proliferation of white blood cells and can be life-threatening if not diagnosed early. Traditional diagnostic techniques rely on manual examination of microscopic blood smear images, which can be time consuming and prone to human error. Recent advancements in deep learning, particularly Convolutional Neural Networks (CNNs), have enabled automated detection and classification of leukemia from microscopic images with high accuracy. This study proposes a deep CNN model trained on a dataset of microscopic blood smear images to distinguish between normal cells and leukemia-affected cells. The model is evaluated based on its accuracy, sensitivity, and specificity, demonstrating its potential as a reliable diagnostic tool. The proposed CNN architecture leverages multiple layers of feature extraction, enabling it to detect subtle morphological changes in the blood cells that indicate leukemia. By employing image augmentation techniques and fine-tuning hyperparameters, the model achieved a high level of precision in detecting both acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML). The results suggest that deep learning models, when integrated into clinical workflows, can significantly enhance early detection, reduce diagnostic errors, and support hematologists in making faster and more accurate decisions.