Web-based Approach for Detection of Acute Lymphoblastic Leukemia From Microscopic Blood Cell Images Using Convolutional Neural Network

Abstract Acute Lymphoblastic Leukemia (ALL) is a blood tumor that affects bone marrow and blood. Leukemia affects young people and adults worldwide. Early diagnosis of leukemia provides a patient with the right treatment option, especially for children. In medical image processing, computational tools play a major role in the field of research. The objective of this paper is to predict the ALL cells using image processing and a Computational Neural Network (CNN). 498 images were collected from the publicly available datasets and the analysis was performed. In the preprocessing stage, noise in the image is eliminated by median and weiner filters and the RGB color leukemia images are converted into lab color images as it enhances segmentation of the nucleus of the image. In this study, the k-means clustering algorithm was applied to segment the nucleus from microscopic blood cell images. Based on the characteristics of the nucleus in the leukemia image, the tumor is detected. CNN takes different features from the input blood sample image and assigns weights to the input image, which helps the network with classification. This model was evaluated with a sensitivity of 98.2%, a specificity of 97%, and an accuracy of 98%. Finally, a flask web page is developed using python to display the output.