Body area networks and remote health monitoring systems allow for collecting physiological data from patients, and provide a platform to utilize analytics algorithms to predict medical conditions. This paper presents an effective predictive analytic approach for hospital readmission prediction for patients with Congestive Heart Failure (CHF) and based on the physiological data collected in last days of hospital stay. We examine the proposed algorithm on the Electronic Health Records (EHR) of UCLA Hospital containing over 10 million clinical measurements collected from approximately 10,000 patients hospitalized at the UCLA Medical Center. The results show that it is possible to predict medically adverse events (e.g. hospital readmissions) for CHF patients if we have access to recent physiological measurements. This study suggests that a remote health monitoring system can provide an effective platform to reduce readmission rates by early prediction of readmissions based on freshly collected data, and then applying appropriate early clinical interventions to prevent the readmission.
|Original language||English (US)|
|Journal||BodyNets International Conference on Body Area Networks|
|State||Published - 2017|
|Event||11th International Conference on Body Area Networks, BODYNETS 2016 - Turin, Italy|
Duration: Dec 15 2016 → Dec 16 2016
- Cognitive Heart Failure (CHF).
- Predictive Analytics
- Remote Health Monitoring Systems (RHMS)
ASJC Scopus subject areas
- Artificial Intelligence
- Computer Networks and Communications
- Computer Science Applications