Wireless Automated Inpatient Monitoring System


Falls are listed as the third most common cause of unintentional injury and death in all age groups and the leading cause of death in adults over the age of 65 (Centers for Disease Control and Prevention Injury Center, 2007). Hospital settings can be dangerous, due to factors such as the unfamiliar environment, change in medical conditions and medication side effects. Currently, inpatient falls occur at a rate of 2.3-7 falls per 1,000 patient days, with an estimated cost per fall equaling at least $6,437.


Figure 1. The wearable node developed for this project

This project aimed at design and development of a wearable device called wireless automated inpatient monitoring system (WAIMS) that can monitor and identify movements of patients in the hospital settings in real-time. If the patient is considered at high-risk of falls, and if they perform certain movements without seeking assistance (e.g., getting out of the bed), the system can notify the care-givers. This project created a complete system that includes wearable nodes, base-stations at the nursing station that provides visual feedback on activities of the patients in real-time and signal processing algorithms for identifying movements. Extensive human subject studies were conducted to validate the system. This includes (N=22) healthy younger subjects at UT-Dallas and UT-Arlington and (N=20) elderly patients at the THR-Plano.


The project has ended, and the operation and the effectiveness of the system were successfully demonstrated; in addition, the outcome of this project has led to an NIH sponsored study. The NIH award is focused on creating a biofeedback system that will measure sway and gait measures and train users to reduce excessive sway when walking, with the goal to reduce the risks of falls. The study was funded for $366K over duration of three years.


Figure 2. The graphical user interface (GUI) on nursing station showing the movements in real-time



Publications, Patents filed, additional funding secured for each project

  • Jerry Mannil, Mohammad-Mahdi Bidmeshki, Roozbeh Jafari, Rejection of Irrelevant Human Actions in Real-time Hidden Markov Model Based Recognition Systems for Wearable Computers, ACM International Conference on Wireless Health, October 10-13, 2011, San Diego, CA.
  • Nimish Kale, Jaeseong Lee, Reza Lotfian and Roozbeh Jafari, Impact of Sensor Misplacement on Dynamic Time Warping Based Human Activity Recognition Using Wearable Computers, ACM International Conference on Wireless Health, October 23-25, 2012, San Diego, CA.
  • The Impact of Vibrotactile Biofeedback on the Excessive Walking Sway and the Postural Control in Elderly, Omid Dehzangi, Zheng Zhao, Mohammad-Mahdi Bidmeshki, John Biggan, Christopher Ray, Roozbeh Jafari, ACM International Conference on Wireless Health, November 1-3, 2013, Baltimore, MD.


TexasMRC grant program provides a unique opportunity to identify projects that are of interest to the health-care providers, industry and academia. This will lead sponsored projects that not only have a strong basic research component; but are also in line with the needs and priorities of health-care providers and industry.

We would like to thank Ms. Lauren Braunfeld, Ms. Ashley Jones, Mr. Daniel Gonzalez, and Ms. Tamara Plant for their efforts and support conducting the human subject investigation at the THR Plano.

Primary Investigators:

Roozbeh Jafari (UTD)
Christopher Ray (UTA)
Mike Motes (UTA)
David Keller (UTA)
John Hart (UTD)



Leave a Comment About This Project

Your email address will not be published.

Please Prove You Are Human *