COVID-19 prediction and symptom analysis using wearable sensors and IoT

The purpose of the research work is to focus on the deployment of wearable sensors in addressing symptom Analysis in the Internet of Things (IoT) environment to reduce human interaction in this epidemic circumstances.

COVID-19 pandemic has distracted the world into an unaccustomed situation in the recent past. The pandemic has pulled us toward data harnessing and focused on the digital framework to monitor the COVID-19 cases seriously, as there is an urge to detect the disease, wearable sensors aided in predicting the incidence of COVID-19. This COVID-19 has initiated many technologies like cloud computing, edge computing, IoT devices, artificial intelligence. The deployment of sensor devices has tremendously increased. Similarly, IoT applications have witnessed many innovations in addressing the COVID-19 crisis. State-of-the-art focuses on IoT factors and symptom features deploying wearable sensors for predicting the COVID-19 cases. The working model incorporates wearable devices, clinical therapy, monitoring the symptom, testing suspected cases and elements of IoT. The present research sermonizes on symptom analysis and risk factors that influence the coronavirus by acknowledging the respiration rate and oxygen saturation (SpO₂). Experiments were proposed to carry out with chi-Square distribution with independent measures t-Test.

IoT devices today play a vital role in analyzing COVID-19 cases effectively. The research work incorporates wearable sensors, human interpretation and Web server, statistical analysis with IoT factors, data management and clinical therapy. The research is initiated with data collection from wearable sensors, data retrieval from the cloud server, pre-processing and categorizing based on age and gender information. IoT devices contribute to tracking and monitoring the patients for prerequisites. The suspected cases are tested based on symptom factors such as temperature, oxygen level (SPO₂), respiratory rate variation and continuous investigation, and these demographic factors are taken for analyzed based on the gender and age factors of the collected data with the IoT factors thus presenting a cutting edge construction design in clinical trials.

The contemporary study comprehends 238 data through wearable sensors and transmitted through an IoT gateway to the cloud server. Few data are considered as outliers and discarded for analysis. Only 208 data are contemplated for statistical examination. These filtered data are proclaimed using chi-square distribution with t-test measure correlating the IoT factors. The research also interprets the demographic features that induce IoT factors using alpha and beta parameters showing the equal variance with the degree of freedom (df = 206).