Blood cancer detection through deep learning
Aakash Gorai, Ranjan Walia, Shweta Agarwal · 2025
Acute Lymphoblastic Leukemia (ALL) is a critical blood cancer that mostly affects children and requires quick and precise diagnosis to achieve the best treatment outcomes. Traditional diagnostic techniques, such as visual inspection of blood smears, are often time-consuming and reliant on individual interpretation. DL techniques have greatly impacted medical diagnostics in the past few years, particularly in the field of image-based disease detection. This study thoroughly examines how deep learning is used in identifying acute lymphoblastic leukemia (ALL), including both convolutional neural networks (CNNs) and hybrid models. We highlight the benefits and limitations of various approaches and tackle challenges such as class imbalance, dataset diversity, and integration into clinical processes with 96% accuracy. Furthermore, we introduce sophisticated deep learning platforms that fix these shortcomings and improve detection accuracy and ability to generalize. The suggested approaches aim to improve the precision, effectiveness, and usability of models in real-world clinical environments.