Abstract:

It is a serious illness caused by a blood parasite called Plasmodium. The standard way to diagnose infusion is the visual examination of blood transfusions to parasites infected red blood cells by specialists under a microscope. This method is inefficient and the diagnosis depends on the experience and knowledge of the person who performs the examination. Previously, automatic image recognition technologies based on machine learning have been applied to the thyroid blood cells for diagnosis. The purpose of this study is to distinguish the spinal cells from non-parasitic or non-infected, a integrated diagnosis system based on pre-trained evolutionary nerve networks has been proposed. In this study, a total of 27558 images were used in the National Institutes of Health (NIH) to classify staminal cells. Experimental studies have resulted in the classification of staminal cells, respectively with the AlexNet, GoogleNet, SqueezeNet, ShuffleNet ESA models 95.77, 96.31, 95.95, 96.44 and 0.9880, 0.9887, 0.9888, 0.9923 EAA performance values. The best outcome in the classification of decomposition cells was achieved using the ShuffleNet CNN model with a 96.44% accuracy value.

Keywords: