Obsolete personal information update system: towards the prevention of falls in the elderly

Falls stand for a prevalent problem among the elderly and a significant public health concern. In recent years, a growing number of apps have been developed to assist in terms of the delivery of more effective and efficient falls prevention programs. All of these apps rely on a massive elderly personal database gathered from hospitals, mutual health groups, and other organizations that help the elderly. Information on an older adult is constantly changing, and it may become obsolete at any time, contradicting what we currently know about the same person. As a result, it needs to be checked and updated on a regular basis in order to maintain database consistency and hence provide a better service. This research work describes an Obsolete Personal Information Update System (OIUS) developed as part of the elderly-fall prevention project. Our OIUS intends to control and update the information gathered about each older adult in real-time, to provide consistent information on demand, and to provide tailored interventions to carers and fall-risk patients. The method discussed here is based upon a polynomial-time algorithm built on top of a causal Bayesian network that models the older adults data. The outcome is presented as an AND-OR recommendation Tree with a certain level of accuracy. On an aged personal information base, we perform an empirical study for such a model. Experiments corroborate our OIUS’s viability and effectiveness.