OPENCARE PERSONAL COMMUNICATION DEVICE


INTRODUCTION

This page discusses the OpenCare Personal Communication Device. A mobile wrist or neck-born, device for monitoring the users vital signs while not at home, monitoring activity (using an accellerometer), allowing the user to communicate using either regular phone numbers, or special internal IP communication, recieving and sending text messages (both SMS and OpenCare Alert and Info messages), while also providing both explicit and implicit fall detection and alert features (detecting when a user is falling). This is work in progress, and we expect a fully functional prototype in 2009. Already, three separate prototypes have been developed by three teams within the OpenCare Project. These have not yet been merged into a single product. The first of these prototypes includes the hand-over sensor communication solution; the second includes support for user friendly phone calls, infrastructure messages (including from the reminder system), and alerts (including GPS support). The third prototype includes the accelerometer and using these data in conjunction with other sensors. These separate products needs to become a unified solution, which is the current focus of the project.

The OpenCare infrastrcuture is devided into tiers (see figure 1). At the Home-tier we find a touch-screen based stationary computer unit, the Basestation (also called BaseTerminal), which samples and distributes data from all sensor types, including for instance ECG, blood-pressure, weight, automatic medicine dispenser units [6] and others. This device is using a plugged power supply configuration, which was deemed necessary, in order to supply the needed amount of power for keeping several different wireless connections active, processing business logic and user interface tasks, as well as maintaining continuous communication with the Central-tier using one or more broadband connections for fail-safe operation.

Other research teams have advocated using a mobile device, e.g. a cell phone, as the primary or even sole communications platform [9, 10, 11]. This approach has various flaws however, including the limited processing power of the device, the limited band-with and number of communication channels (for instance limiting Bluetooth communication to one sensor at a time, not allowing for ZigBee and proprietary protocols etc). The most significant shortcoming of the mobile device approach however, is the limited battery resources. Our measurements indicate that it is not possible to maintain even one day of continuous communication with a Bluetooth-based ECG device. This is of course not in any way acceptable for any kind of in-situ deployment, as the battery should be charged or changed at an unacceptable frequency. Of course, some sensor types might only need to be read at a very infrequent interval, which of course might help the expected battery life expectancy to become more reasonable.

systemOverview_small.PNG
Figure 1. The four tiers of the OpenCare Infrastructure [8], along with the major components contained within them. In the Mobile-tier, we find the Personal Communication Device, including easy cell phone access, Bluetooth sensor communicaiton capabilities, GPS and accellerometer for movement and location tracking.

However, it seems evident that a mobile platform is not applicable for a generic platform such as the OpenCare Infrastructure, where many different types of sensors might be applied, and where the sensor equipment vendors might have different communication frequency and band-with needs. Here we need to provide a stable and effective platform.

Of course the stationary platform comes at a cost, namely that of less mobility for the user. The user cannot leave his or her home and be under continuous observation with the present solution.
Thus the current problem facing the OpenCare Infrastructure is the lack of mobility and thus flexibility for users leaving their home. The solution to this is to augment the OpenCare Infrastructure with a device with mobile capabilities.

The Personal Communication Device is a mobile extension to the existing infrastructure. It is situated in the Mobile-tier. It is simply a combination of a very non-complex interfaced cell phone optimized for elderly and cognitive impaired users (see figures 4), combined with a sensor gateway functionality, an accelerometer as well as a GPS service for location tracking. The device utilizes knowledge of whether the device (and thus the user) is currently at home and therefore within range of the stationary Basestation, or not. At a regular interval the device will check for this. In case the device is at the user’s home and within range of the Basestation, it will regularly check for new sensor device drivers on the Basestation, and automatically synchronize compatible drivers. A sample scenario might be, that a new Bluetooth based ECG device has been brought to the users home, e.g. to replace a less accurate or outdated device. The Basestation will automatically discover this new device through its zero-configuration capabilities [7], download the device drivers for it, and instantiate communications. Within 10 minutes, the Personal Communication device would have checked with the Basestation for new compatible host drivers, it would have found the new ECG device, and downloaded the drivers for it, but it would not yet have started communication (as the Basestation is currently taking care of this). Once the user leaves his home, the Personal Communication Device will simply perform a take-over of communication with all compatible sensors, including the newly discovered ECG device, and thus the user will be monitored by the Personal Communication Device until returning home, where the Basestation again will assume communications, and thus save the battery resources of the Personal Communication Device (see figures 2 and 3 for the different deployment scenarios).

pcd1.PNG
Figure 2. Deployment diagram of the Personal Communication Device (PCD) communicating with the Basestation using Bluetooth. This is both used for synchronizng drivers, checking for location (at home or not), and sending data to the Central-tier.

pcd2.PNG
Figure 3. Deployment diagram of PC communications when not in the users home. Here we use the GSM network (or other cell phone networks available) to communicate directly with the Central-tier.


Besides acting as a mobile extension for monitoring and distributing data from the sensors, the device is in itself a sensor, keeping track of the user’s movement and activity level using an accelerometer. It also serves as a communications platform, where information, including amongst other reminders and status messages, may be sent to the user wherever he may be located.

pcd3.PNG
Figure 4. The Personal Screen-shot of the PCD touch-screen, frontpage (default view). Besides showing the time, there are only two menu options for the user. Alert or “Ring”(make a phone call), making the user interface well suited for cognitive impairred users and the elderly.

The device also features voice phone capabilities, with an extremely simple user interface for cognitive impaired users (see figures 4 and 5). Finally, the device features an alert feature, where the user might send an alert in case of getting lost, or having a problem (see figure 4). In case the user is at home (and thus within range of the Basestation) the alert will send the home address (via the Central-tier) to the emergency call-center, and if not, a GPS-reading is attempted, and converted to an address if possible.

pcd4.PNG
Figure 5. The phone interface (after choosing “Ring”), is extremely simple. The user only has to touch the name that he wants to call, and the call is placed. Instead of using text buttons, images (e.g. with faces) may be used instead.

pcd5.PNG
Figure 6. A screen-shot from the PCD. Here a message from the Reminder service, reminding the user to “take his pills”.


TECHNOLOGIES

The PCD is based on the Windows Mobile platform running .NET Compact Framework, using the C# programming language. The use of SOAP based web services as the primary communication standard ensures a degree of platform independence, including the possibility to use the PCD with other infrastructure systems than the OpenCare Infrastructure, but also allowing for replacing the Windows Mobile platform with another embedded platform.

FUTURE WORK

As discussed above, three separate prototypes have already been developed by three teams within the OpenCare Project. These have not yet been merged into a single product. The first of these prototypes includes the hand-over sensor communication solution; the second includes support for user friendly phone calls, infrastructure messages (including from the reminder system), and alerts (including GPS support). The third prototype includes the accelerometer and using these data in conjunction with other sensors. These separate products needs to become a unified solution, which is the current focus of the project.

The PCD hardware needs to be further adopted and customized to reduce the size and battery demands. Also, a prototype project trying to equip the PCD with wireless power (for easy and wireless recharging) is currently in progress. A stand-alone working prototype has been constructed.
Finally, it is important to mention that end-user testing has not yet commenced with the system. Usability being a major design criteria for the PCD, this is also of major importance, and is scheduled to commence as soon as a unified prototype is complete.

REFERENCES

  • [1] S. Wagner, "Towards an open and easily extendible home care system infrastructure”, Proceedings of the 2nd International Conference on Pervasive Computing Technologies for Healthcare, Tampere, Finland, 2008
  • [2] J.E. Bardram, A. Mihailidis, D. Wan, “Pervasive Computing in Healthcare”, CRC Press, 2006.
  • [3] H. Kautz, L. Arnstein, G. Borriello, O. Etzioni., D. Fox. “An overview of the assisted cognition project.”, AAAI-2002 Workshop on Automation as Caregiver: The Role of IntelligentTechnology in Elder Care, Edmonton, Alberta, 2002.
  • [4] A. Bamis, D. Lymberopoulos, T. Teixeira, A. Savvides, “Towards precision monitoring of elders for providing assistive services”. In PETRA '08: Proceedings of the 1st international conference on Pervasive Technologies Related to Assistive Environments, pages 1-8, New York, NY, USA, 2008. ACM.
  • [5] US Census Bureau, International Database. Accessed via Internet: http://www.census.gov/ipc/www/idb/, November 2008.
  • [6] S. Wagner, “Zero-configuration of pervasive healthcare sensor networks” Proceedings of the The Third International Conference on Pervasive Computing and Applications (ICPCA2008), Alexandria, Egypt, 2008.
  • [7] R.A. Soerensen, J.M. Nygaard, “Distributed zero configuration base station”, Proceedings of the 2nd International Conference on Pervasive Computing Technologies for Healthcare, Tampere, Finland, 2008
  • [8] C. Nielsen, S. Wagner, “OpenCare Project: An Open, Flexible and Easily Extendible Infrastructure for Pervasive Healthcare Assisted Living Solutions”, Proceedings of the 3rd International Conference on Pervasive Computing Technologies for Healthcare, London, United Kingdom (2009).
  • [9] H.J. Lee, S.H. Lee, K. Ha, H.C. Jang, W. Chung, J.Y. Kim, et al., “Ubiquitous healthcare service using Zigbee and mobile phone for elderly patients”. International Journal of Medical Informatics 2009 3;78(3):193-198
  • [10] R. Lee, C. Hsiao, C. Chen, M. Liu, ”A Mobile-Care System Integrated with Bluetooth Blood Pressure and Pulse Monitor, and Cellular Phone”. IEICE Trans.Inf.Syst. 2006 /5/1;E89-D(5):1702-1711
  • [11] A. Lorenz, R. Oppermann, “Mobile health monitoring for the elderly: Designing for diversity”. Pervasive and Mobile Computing ;In Press, Corrected Proof., 2008
*