This document describes a method for detecting and classifying leukemia using microscopic blood smear images. The method involves preprocessing images using techniques like median filtering for noise removal. Images then undergo segmentation using grayscaling, binarization, and adaptive thresholding. A convolutional neural network (CNN) is trained on a dataset of labeled images to classify images as acute, chronic, or normal leukemia. The CNN model provides accurate classification of leukemia types from blood smear images. Additionally, the results can be stored on a server using internet of things (IoT) for remote monitoring of patient data.