Smart wearable Cardio-health monitoring system using Deep Learning Technologies

Abstract

 Every year, the proportion of the population suffering from Cardiovascular Disease (CVD) rises dramatically. According to the World Health Organization (WHO), this disease kills millions of people each year, which is heartbreaking. The remarkable improvement in wearable technology has created opportunities to provide numerous intelligent approaches for tackling this condition efficiently. Furthermore, early detection of CVD improves medication and speeds up the treatment process by clinical professionals. The seriousness of this issue motivated us to propose a wearable smartwatch by integrating Deep Learning (DL) and Internet of Things (IoT) technologies. The DL model is designed using transformer encoders for cardio-health status prediction. To achieve this, Electrocardiogram (ECG) data from the MIT-BIH database is used. The effectiveness of the model is evaluated in terms of accuracy and execution time. Additionally, the model's output is compared with the Convolutional Neural Network (CNN) model. The proposed model achieves the highest accuracy of 98.04% on 2500 test samples. The proposed model is deployed in the cloud. The ECG sensor is fixed to the watch to collect ECG signals from the person and send them to the cloud. The cloud analyzes the data using the deployed DL model and predicts the cardio-health status. If the cardio-health is abnormal, the cloud sends an immediate alert to the registered mobile number. The proposed smart wearable watch can help individuals monitor their health and improve their quality of life.