چكيده به لاتين
Ongoing changes in world demographics and the prevalence of unhealthy lifestyles are imposing a paradigm shift in the healthcare landscape. Nowadays, cardiovascular disease is the first cause of death in the world. The Electrocardiogram (ECG) signals play a significant role in the analysis of the electrical behavior of the heart and the diseases of this vital organ of the body. Hence, many studies are devoted to obtaining heart signals, processing them, and analyzing cardiac behavior. On the other, one way to reduce cardiovascular disease is to make an intelligent life. Wireless Sensor Networks (WSN) play a role in this regard. Patients using this network can send their biological signals (such as electrocardiogram) to doctors or the nearest health clinics in their homes.
The limitation in the energy consumption of wireless sensor nodes has prevented these devices from playing their full role. To reduce the energy consumption of sensor nodes, signal processing techniques on nodes have been used simultaneously to obtain Electrocardiogram signals from the patient's body.
In this project, for the first time, we have analyzed the Electrocardiogram signals and classification using Recurrent Neural Networks(RNN). According to the proposed method in this study, by using MATLAB computational software, the classification of healthy and patient signals was 94.8%. Also, regarding the diagnosis of patient's heart signals, the best network trained in this project was 98.4% correct.
This project has shown that the reduction of energy consumption in the transmission of data from the wireless sensor sensor nodes has reached 62.7%.
