Efficient Approach for Patient Monitoring: ML-Enabled Framework with Smart Connected Systems

Healthcare professionals are required to adhere to strict precautions. Thus, continuous monitoring, compliance with safety standards, and prompt care are essential. Hospitals employ various methods and devices to ensure the patient’s well-being. However, many of these efforts are compromised when the observation is neglected. This paper presents an effective approach to enable doctors (or) supervisors to track and forecast the patient’s health condition during the treatment. The approach leverages the concept of Internet of Things (IoT) for seamless data acquisition, data analytics for visualization, and machine learning (ML) to train models with the acquired data for future prediction of similar conditions. For monitoring, sensors are used to collect data such as the ambient state and the location to verify that the patient (or) person under observation is within the expected range using range detection techniques supported by Bluetooth master-slave communication. Computations are performed in the backend such that the alerts are notified based on the conditions assigned to the respective patients. In case of emergency, we can reliably predict the condition of a patient with improved accuracy. Moreover, the stored data is fed to an ML framework for data training and ML modeling. Therefore, this design can serve as an optimal model to address the much-needed advancement in healthcare.