The development of an online pneumonia risk prediction system

Pneumonia disease is a lung inflammatory condition that affects 7% of the world population, causing 4 million deaths yearly. The risk prediction model is an effective way to identify early risk for individual patients, to assist the clinician in making decisions. Many pneumonia prediction systems have been developed. However, the existing models focus on predicting mortality in hospitalized patients, predicting the risk in the stroke patient, prediction of risk of ICU admission in patients with pneumonia, and the risk in older adults. By identifying the risk factors contributing to pneumonia, early prevention can be made to improve health life factor. This study was done to design new algorithms for pneumonia risk prediction system based on the pathway of the infection through eight human body system. The risk was identified from the human body systems and classified into three categories such as molecular structure (Low Energy Level (LEL)), physiological measure (Middle Energy Level (MEL)), and bioenergy symphony (High Energy Level (HEL)). In this study, new risk factor categories which is HEL factor have been implemented. The model was verified using 20 pneumonia patients and 20 healthy subjects between 20 to 30 years old ages. The study showed that the accuracy of the prediction system is 73% with 50% sensitivity and 95% specificity.