White Blood Cells Subtypes Classification Using Fast Traditional Convolutional Neural Network

In the medical field, detection and classification of white blood cells are essential. This analysis can be constructive when several diseases and infections are detected and treated. The purpose of this study is to develop an image classifier capable of automatically classifying the image of blood cells into four white blood cell subtypes: the cell neutrophil, eosinophil, lymphocyte, and Monocyte. A Fast Traditional Convolutional Neural Network (FTCNN) model has been designed to create an image classifier, which ensures high efficiency and accuracy when working with a large number of images. The proposed model has been trained on about 12,515 subtyped blood microscopic images and provided 98.23% accuracy and 84.64% accuracy, respectively, during training and testing. To help the medical department combat WBC-related diseases, this blood cell classification model can be used as a base for establishing various diagnostic systems.