What is Mobility?
Mobility can be defined as the ability to move independently from one point to another. For people with physical disabilities, mobility deficits negatively impact quality of life and health status. For this population, mobility includes walking on various terrains, using a wheelchair, driving, negotiating stairs and ramps, etc. Optimizing mobility enhances independence, reduces isolation, promotes participation, and improves quality of life.
Why monitor or measure mobility?
Understanding a person’s mobility capacity is important for healthcare providers since mobility status assessments can affect clinical decision-making, assistive device prescription, environmental modification decisions, and other social and physiological factors. Self-reporting or assessment in the clinic can be poor methods for understanding how a person moves in the community. A better understanding of a patient’s ‘real life’ mobility can result in better patient care.
The most appropriate and objective way to evaluate a patient’s mobility is by making unobtrusive measurements as the person proceeds through the activities of daily living. This can be achieved with a wearable mobility monitoring system (WMMS) that the person wears while going about their daily life. Movement is logged throughout the day and thus provides the healthcare provider with ‘real-life’, objective data on the person’s mobility.
The objective of this project is to use the BlackBerry platform to develop a Wearable Mobility Monitoring System (WMMS) that monitors a person’s mobility in a community (non-clinical) environment and then supplies the healthcare provider with the relevant objective data.
Our first-generation Wearable Mobility Monitoring System
Our recent research has demonstrated the viability of using Smartphones, particularly the BlackBerry platform, as the basis for a WMMS. A prototype WMMS was developed that consisted of sensors integrated into the phone’s holster (SmartHolster), GPS from the phone, phone camera image capture, and custom software. Various mobility states were defined such as, sitting, standing, walking, climbing stairs, etc. The system continuously monitored the person and attempted to identify activity changes-of-state while wearing the WMMS. Any detected change-of-state was identified in real time (e.g. sit-to-stand) and the WMMS took a photo whenever a change of state was detected. The digital photo that accompanied the change in state was stored so that the context of the motion could be identified during post-processing.
Thus, the real-time signal processing tasks were to identify changes-of-state, capture and log an accompanying photo, and categorize the activity. Post-processing consisted of downloading the logged data and a human operator identified the mobility context from the logged photos and sensor data. The software’s accuracy for identifying various states was verified by taking an accompanying video as subjects performed a pre-set sequence of mobility tasks.
This work is documented in two conference papers, one journal paper and one master’s thesis (which was nominated for uOttawa Master’s Thesis Prize, Sciences Division).
- G. Hache, E. Lemaire and N. Baddour, Development of a Wearable Mobility Monitoring System, Proceedings of the Canadian Medical and Biological Engineering Conference, Calgary, May 2009. Paper
- G. Haché, E. Lemaire and N. Baddour, Mobility Change-of-State Detection Using a Smartphone-based Approach,Proceedings of the IEEE International Workshop on Medical Measurements and Applications, pp. 43-46, May 2010, Ottawa. Paper
- G. Haché, E. Lemaire and N. Baddour, Wearable Mobility Monitoring using a Multimedia Smartphone Platform, accepted for publication in IEEE Transactions on Instrumentation and Measurement, 2010. Pre-print and IEEE print.
- Gaetanne Haché, M.A.Sc., 2010 (co-supervision with Dr. Ed Lemaire), Development of a Wearable Mobility Monitoring System. This thesis was nominated for UOttawa Master’s thesis prize, Science Division.
Second generation WMMS
Based on the success of the first generation WMMS, a second generation WMMS was developed with the goal of using only the BlackBerry’s built-in, onboard hardware and without requiring the use of the Smart Holster or its sensors. This was the focus of (another) master’s thesis project which was completed in November 2011. So far, two conference papers, one thesis and a journal paper describe these efforts, with more forthcoming.
The preliminary results of this work were presented in a poster at RIM Research Day in Waterloo, December 2010.
So far the work has been presented at two conferences and also in one journal publication:
- H.H. Wu, E. Lemaire and N. Baddour, Change-of-State Determination to Recognize Mobility Activities Using a BlackBerry Smartphone, Proceedings of the 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC ’11), Boston, 2011. Paper
- H.H. Wu, E. Lemaire and N. Baddour, Using the BlackBerry to Assess Mobility for Rehabilitation, Proceedings of the Canadian Medical and Biological Engineering Conference, Toronto, June 2011. Paper
- H.H. Wu, E. Lemaire, N. Baddour, Activity Change-of-State Identification Using a BlackBerry Smartphone, Journal of Medical and Biological Engineering, Vol. 32, No 4, pp. 265-272, 2012. Read Online.
Work on this project is ongoing. The latest work is being presented at the University of Ottawa Graduate Student Poster Competition. Check out Marco Tundo’s poster.
We now have 2 apps on BlackBerry Appworld:
- TOHRC Data Logger, an easy to use tool for capturing sensor data from your BlackBerry and saving this information to the memory card. Download from BlackBerry AppWorld.
- TOHRC WMMS Acc: TOHRC Wearable Mobility Monitor (WMMS) uses your BlackBerry’s accelerometer and GPS to determine the movement activities you are performing (walking, sitting, lying down, etc.). Download from BlackBerry AppWorld.
Third generation WMMS
The second generation WMMS focussed on using only onboard hardware and some of the algorithms had to be revised to deal with the limited data rate of the Smartphone we were using at the time. The third generation WMMS was again redeveloped, this time to run on BlackBerry’s new BlackBerry 10 phone, featuring much more powerful hardware. Also, we have been working on trying to use the video collected by the phone to aid in identifying context (stairs vs flat ground) which are sometimes difficult to distinguish using signal processing alone. These latest efforts are documented in two conference papers and one MASc thesis:
- P. Moradshahi, J. Green, E. Lemaire, N. Baddour, Differentiating Two Daily Activities Tthrough Analysis of Short Ambulatory Video Clips, Proceedings of the IEEE International Symposium on Medical Measurements and Application, Gatineau, Canada, May 2013. Read Online.
- M. Tundo, E. Lemaire, N. Baddour, Correcting Smartphone Orientation for Accelerometer-Based Analysis, Proceedings of the IEEE International Symposium on Medical Measurements and Application, Gatineau, Canada, May 2013. Read Online.
- Marco Tundo, M.A.Sc., 2014 (co-supervision with Dr. Ed Lemaire), Development of a Human Activity Recognition System using Inertial Measurement Unit Sensors on a Smartphone.
Interested in this project?
We still have some some funding to continue this project. We are currently working on context identification from video and also improvements to the WMMS itself, in addition to a lot of “back-end” programming that needs to be done.