Robust preprocessing for health care monitoring framework

Remote health care monitoring is an emerging application that helps to reduce the cost of health care and at the same time improve its quality. However, by its nature, medical sensor data is often unreliable and massive as the data is collected from numerous sensors operating in noisy environments. Therefore, ensuring the reliability of the sensor data and the scalability of the health care monitoring services are among major challenge and is a determining factor in the success of the system. In this paper, we propose a robust and flexible preprocessing module as part of an active health care monitoring framework. It is responsible for preparing the sensor data and performing some initial assessment of the data for input to later modules. The proposed preprocessing architecture contains five stages: validation, transformation, cleaning, reduction and cross-verification. The module is evaluated using synthetic blood pressure data.