Utilizing Artificial Intelligence and IoT Technologies for Enhanced COVID-19 Diagnosis

The rapid spread of COVID-19 and the high mortality toll globally necessitated an immediate reaction from several industries to provide an early illness forecast. As a result, the Internet of Things and artificial intelligence techniques are crucial for identifying and predesignating the illness so that patients can receive the proper medication and care as soon as possible. To accomplish the above-mentioned goal, several datasets are examined and categorized in this work, considering different patient characteristics, including gender, age, diseases, and obesity. The study will show that multilayer perception and logistic-based classification algorithms provide the highest accuracy compared to other algorithms. A dataset of 1217 records of patient data with 37 attributes is considered to cover all possible patient characteristics, such as patient information (age, sex, region, etc.), symptoms (fever, coughing, sore throat, temperature, etc.), risk factors (diabetes, asthma, liver disease, etc.), and lab values (CT scans, serum levels, X-ray reports). According to the experimental results, the multilayer perception algorithm has the highest accuracy of 96.27%. This result reveals that using artificial intelligence can be beneficial for the early prediction and diagnosis of COVID-19 disease, and hence, effective and proper treatment for patients can be considered promptly.