Separation of white blood cell types for disease diagnosis using artificial intelligence techniques

White blood cells are vital cells that protect the body against infections, foreign substances, and other harmful threats, forming a fundamental component of the immune system. Produced in the bone marrow and distributed throughout the body via the bloodstream, these cells play a key role in initiating, regulating, and implementing defense mechanisms against pathogens. Therefore, the automated detection of white blood cell types is of great importance. In this thesis, a hybrid model was developed for the automated detection of white blood cells using Convolutional Neural Networks (CNN) and Vision Transformer (ViT). First, feature extraction was performed using 13 different CNN and ViT architectures. Then, the two models that produced the most successful results formed the basis of the proposed model. Feature representations obtained from the two best-performing models were combined, and then a VOTE-based decision mechanism was applied. As a result, an accuracy rate of 97.44% was obtained. The proposed model was also compared with the other models used in the study. As a result, it was observed that the proposed model produced more successful results than the other models. Therefore, we believe that the proposed model can be used in the detection of white blood cells. The proposed model will reduce the workload of specialists and shorten the diagnosis and treatment time. Consequently, patients can begin treatment earlier.