White Blood Cell Classification using Deep Learning

In this paper, a hybrid Convolutional Neural Networks (CNNs)-based system for classifying white corpuscle from blood smear images is presented. The major objective is to effectively classify various types of white corpuscle using CNNs, which will help with early disease identification and medical diagnosis. The suggested method entails gathering a varied dataset of blood smear images, standardizing them through preprocessing, and then using data augmentation approaches to improve model generalization. On the dataset, various CNN architectures are investigated and tweaked for better classification performance, including pre-trained models like ResNet-50and MobileNet-V2. The research places a strong emphasis on thorough evaluation utilizing a variety of measures and comparison with established image processing approaches using deep learning techniques.