Effect of Image Resolution on The Classification of Cells in Peripheral Blood Smear

Abstract

In recent years, there has been a rise in the prevalence of the practise of applying deep learning to the process of pathological inspection. Deep learning has been the focus of a substantial amount of study that has been carried out in an effort to develop effective treatments for problematic circumstances. However, there is a huge problem in the field of artificial intelligence, and that is the optimization of efficiency at the price of accuracy. This is a problem since it leads to less reliable results. This problem has an immediate bearing on the pathological investigation. Through the use of the cell morphology image data-set, the objective of this study is to locate the optimal balance point between performance and accuracy in the context of the trade-off between the two. Because we need to inspect the cells at a microscopic level, the pictures that are acquired by a microscope are often rather large because we want the images to be of a good quality. Before being put to use for detection, memory errors need to be disassembled first into the component parts from which they are constructed so that they may be avoided.