Prediction of personalised life expectancy using personal health devices in mHealth networks

The ability to predict life expectancy (LE) for an individual or a group of people has been in demand for long, however the accuracy and validity of results are difficult to enhance due to the numerous variables required for consideration. The main causes of issues are that human behaviour and activities can be so different and unpredictable that it is almost impossible to measure, classify, define and predict against generic statistic values, which themselves are too numerous in variables to determine. However, health-related data are becoming increasingly available with the emergence of data science technologies and there has been an increase of smartphone and wearable device applications that allow for health and fitness tracking to aid these demands. Some health-related data, such as calorie expenditure and sleep and heart rate monitoring can be provided by apps that are collected by sensors and processed in the cloud. A personalized life expectancy (PLE) can be provided for individuals to improve wellbeing and encourage healthy lifestyle changes. There is currently no work that has addressed a PLE information that can be customized for the individual. This article proposes a novel and innovative idea of calculating and predicting LE. This paper provides a solution that improves the accuracy of a group LE based on individual health data as well as encouraging individuals to change their lifestyle by monitoring their own PLE to improve their quality of their life.