AN INTELLIGENT FAULT DETECTION FOR IoT-EMBEDDED VITAL MONITORING IMPLANTS USING PRM

Abstract

The Internet of Things (IoT) - embedded vital monitoring revolutionizes the healthcare industry by enabling the personalized, continuous time and data driven patient care system. The incorporation of the IoT-embedded in the medical systems, promotes the early detection of disease, enhances the patient outcomes, and reduces the cost along with the enhanced level of accuracy by marking a significant shift towards the preventive and precision medicine. The major challenge in the IoT-embedded vital monitoring system is the fault occurrence, leading to critical concerns affecting the health of the patients, device functionality and the system performance. The fault in the system leads to increase in False Positive (FP) and False Negative (FN) rate leading to inaccurate data reading leading to failure in personalized treatment plans. To overcome this concern, this research work of detecting the fault in the IoT-embedded vital monitoring system using the hybrid Pearson Correlation, Random Forest (RF) classifier and the Multi linear Regression (MR) technology. The proposed work is trained using two different datasets namely the Kaggle dataset and PhysioNet dataset. Both the datasets are the open access datasets and the performance of the proposed work was compared with the performance of the state of the art methodologies.