Follow-up of patients Using Artificial Intelligence During The Pandemic and Its Application In The Diagnosis of Leukemia

Elif GÜLER KAZANCI; Emine BÜYÜKKAYA

Follow-up of patients Using Artificial Intelligence During The Pandemic and Its Application In The Diagnosis of Leukemia

Authors : Elif GÜLER KAZANCI, Emine BÜYÜKKAYA

Pages : 54-63

View : 66 | Download : 13

Publication Date : 2021-09-30

Article Type : Research Paper

Abstract :The prestige of the recent technologies such as “deep learning”, “internet of things insert ignore into journalissuearticles values(IoT);”, “cloud technology”, “big data”, “machine learning” are increasing day by day, and “artificial intelligence insert ignore into journalissuearticles values(AI);” is one of the most important components that pave the way for the development and transformation of those technologies. Recently, the use of artificial intelligence in medicine has become gradually widespread, and “artificial intelligence technologies” can be used in various fields of medicine today. Similarly, computer aided diagnosis insert ignore into journalissuearticles values(CAD); methods have been employed more and more extensively in medicine. A variety of “machine learning algorithms” have also been generated to diagnose several diseases such as “leukaemia”. As it is well-known, a fast, safe and accurate early-stage diagnosis of leukaemia is a crucial factor in healing patients and saving their lives. The current study, which is based on a method of artificial intelligence, aimed to investigate the use of internet of medical things insert ignore into journalissuearticles values(IoMT); to enable and improve a rapid and safe diagnosis of leukaemia. In the proposed IoMT system, medical devices and tools are connected to network with the help of cloud computing. In this system, the techniques used to identify the subtypes of leukaemia were “residual convolutional neural network” insert ignore into journalissuearticles values(ResNet-34); and “dense convolutional neural network” insert ignore into journalissuearticles values(DenseNet-121);. By means of data magnification techniques, ResNet-34 and DenseNet-121 were both put into service to process multiple image patterns. For all healthy cases, the estimation accuracy of ResNet-34 and DenseNet-121 was measured as 100% and the precision rate was found to be 100%.
Keywords : artificial intelligence, leukemia, machine learning, computer aided diagnosis CAD,

ORIGINAL ARTICLE URL