Leukemia is a blood cancer that has two main types: acute and chronic. Diagnosing acute lymphoblastic leukemia (ALL) can be challenging. However, the subgroups for each type are lymphoid and myeloid, making four different kinds of leukemia in total. To enhance the data, seven distinct image transformation methods were used for data augmentation. A CNN framework was developed to identify all subtypes of leukemia. A multiple type detection system for acute lymphocytic leukemia blood cancer was proposed, which uses a Convolution Neural Network (CNN) model to detect a subtype of ALL through the input provided by the user in the form of a blood cell image. CNNs are now widely used for analyzing medical images, but the classical models require large image databases to achieve high accuracy. To address this problem, a powerful deep CNNs approach was proposed, which produces more precise ALL detection and improves the accuracy of the model using a large dataset. The system generates a report of the overall result, providing the user with information about the blood cancer and precautions to be taken for a particular type of blood cancer.