Intelligent Virtual Ambulance Model using Predictive Learning

Most people are unable to identify their actual illness, and many patients die before they are hospitalized. Therefore, in this research, a health monitoring system (HMS) is proposed and it suggests diseases such as arrhythmia, diabetes, fever, etc. and diagnoses chronic diseases in the ambulance itself and provides real-time prediction for critical and normal patients. The proposed system has a multi-tiered structure where physiological features are captured by wearable medical sensors (WMS) and machine learning suites are used to analyze data and to detect chronic diseases. Results obtained on the real datasets puts forward that sensor data and machine learning can effectively predict chronic diseases in real time. HMS can predict accuracies of various conditions such as arrhythmia (98.89%). Diabetes (96.74%), hypertension (98.85%), fever (99.85%) and thyroid (97.4%). In conclusion, we can say that the proposed HMS can assist us to monitor the affected person's physical condition at his/her residence, or at the ambulance within the golden hour of a patient's life.