ABSTRACT
Acute lymphoblastic leukemia (ALL), a type of blood cancer, predominantly affects lymphocytes, which are specialized white blood cells crucial for immune defense. Swift identification and treatment are critical in battling this illness. Conventionally, leukemia diagnosis has depended on invasive methods such as bone marrow biopsies and lumbar punctures, which, while effective, are also expensive and cause discomfort. Nonetheless, recent advancements in deep learning algorithms have transformed diagnostic methods, boosting both efficiency and accessibility. The suggested Region-Based Convolutional Neural Network (RCNN) plays a significant role in this advancement by utilizing Gaussian filtering and data augmentation strategies to enhance the dataset, resulting in superior training outcomes. Furthermore, it categorizes the Acute lymphoblastic leukemia dataset into specific stages like Benign, Pro, Early, and Pre, and provides an intuitive interface for implementing the trained R-CNN model.
