Have any of your parents, family
members or friends fallen in their homes? Have they
stumbled over a pet, tripped on a throw rug, or slipped
in the bathroom? Falls are especially common amongst
the elderly as they go through a physical transition
period such as slower reactions, changes in gait and
balance, and reduced muscle strength. They would also
suffer from chronic health conditions such as high blood
pressure, heart problems and medications; as well as
environmental hazards such as slippery or uneven floors,
poor lighting, loose rugs and unstable furniture
Statistics has shown that between 1992 and 1993, 17.2%
of Singapore elderly age 60 years and above, have had
at least one fall while one-third of them have had recurring
falls. It has been uncovered that fall-related injuries
have serious repercussions. The elderly would frequently
suffer from severe physical injuries such as fractures
and suffer from psychological damage due to the loss
of self-esteem and fear of falling. In many cases, they
would be admitted into nursing homes due to their inability
to care for themselves. Besides physical and mental
damage, falls such as this are also costly. US statistics
has shown that health care costs for falls and rehabilitations
can come up to 70 billion dollar a year.
I²R has taken up task of coming up with a technology
to assist the elderly in the prevention of falls. Its
success will aid in the individual wellness, social
and economy aspects of the society, thus preserving
and improving the quality of life in the individual
and family. This could potentially lead to long-term
health care benefits and reduction in costs for the
general population.
The I²R fall detection technology operates through
a network of overhead video cameras which are strategically
installed to continuously monitor areas such as the
hospital ward. This technology would not require physical
monitoring as it employs cutting edge computer vision
algorithms to monitor the activities of the elderly.
This is done via programming a number of reliable and
tested attributes that model typical traits of falls
to the programme. The system would then detect the situations
and immediately alert the relevant parties.
Figure below shows the architecture of I²R’s
fall detection system.
Our research efforts have led to several patent-pending
technologies. They are:
- Converting 2D positions to 3D positions.
- Recorded video footages will only give 2D information
of a monitored patient who is in a 3D space. This
technology will map 2D coordinates into 3D space,
thus enabling a more reliable detection of fall incidents.
- Detecting 2D location and Orientation information
of the Head and Body.
- Based on the 2D information and the 2D-3D conversion
technology, we have developed the capability to detect
and analyse sub-actions that could lead to a fall.
- Efficient Camera Calibration for fall detection
- As the existing camera calibration procedure causes
much problems for the system deployment, we have proposed
a fast and easy way that allows the camera calibration
to be done by using a simple tool
Creating Impact
We demonstrated our fall detection technology at the
Silver Industry Conference and Exhibition (SiCEX) 2008
to great success as we grasp the attention of companies
in the healthcare industry. We are currently pursuing
technological trials at hospitals and home environments.
The pictures below depict the successful detection
of a collapse event in a room environment.
For enquiries or explore collaboration, please contact:
Industry Development Department
Tel: 65 6874 8399
Fax: 65 6775 9923
Email: inddev@i2r.a-star.edu.sg
|
