Falls are a significant cause of death and serious injury and result in significant health-care costs. Individuals with a lower extremity amputation due to vascular disease are overwhelmingly elderly (at least 65 years of age) and are at especially high risk of falling. Successful fall prevention strategies depend on understanding how, why, when, and where individuals fall, and what types of falls (e.g., trip, slip, or lateral fll) are likely in a given population. Most studies on falls in amputees to date have relied surveys or questionnaires that are often completed a significant time after the fall and thus rely both on the individual's ability to remember the details of their fall and their willingness to be objective abut how and why they fell. Such approaches are susceptible both to inaccurate memories of the fall and to recall bias-for example, due to embarrassment about falling- and are especially unreliable in the elderly amputees. Mobile phones provide a simple, cost-effective method for detection and characterization of falls. Most available smart phones today have a tri-axial accelerometer, which provides highly accurate fall detection in real-time. Other available applications (or apps) can provide data on activity (running, walking etc.) and environment-such as the weather conditions or population density-that may have contributed to the fall and can pin-point the location of the fall-using GPS technology and highly accurate maps. Mobile phones also have inbuilt data storage and transfer capability, allowing for real-time acquisition and transmission of data. Additionally, mobile phones provide a simple means to contact the individual immediately after a suspected fall to confirm details of the fall (and to ascertain the need for medical assistance). Because mobile phone use is so widespread, there is no stigma associated with carrying such a device, which is likely to lead to high compliance. This study aims to use a mobile phone-based fall detection system in dysvascular amputees to detect falls, characterize the type of fall, analyze environmental conditions that may have contributed to the fall, and determine the longer-term consequences of each type of fall. Data acquired may be used to improve rehabilitation protocols or design better prostheses in order to prevent falls. This technology is ultimately transferrable to many populations with a high risk of falling-for example, the elderly, stroke survivors, or those with other musculoskeletal disorders or disabilities-leading to the design of specific fall prevention strategies for those populations.

Public Health Relevance

Successful fall prevention strategies for the elderly dysvascular amputee population-including better prosthesis design and improved rehabilitation/fall prevention strategies-depend on understanding how, why, when, and where such individuals fall, and what types of fall occur (e.g., trip, slip, or lateral fall). Mobile smartphones with built-in triaxial accelerometers can accurately detect and classify falls, in addition to providing other applications (or 'apps') that can identify contributory environmental conditions (e.g., weather or traffic) or activities (e.g. walking or running) that have may contributed to the fall. Combined with server side analysis of wirelessly transmitted phone data-using machine learning techniques-mobile smartphones provide a simple, portable fall-detection system that generates real-time information on the mechanisms, contributory environmental factors, and consequences of falls in elderly amputees-or in any population at risk of falling.

National Institute of Health (NIH)
National Institute of Biomedical Imaging and Bioengineering (NIBIB)
Research Project (R01)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Lash, Tiffani Bailey
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Rehabilitation Institute of Chicago
United States
Zip Code
Lonini, Luca; Reissman, Timothy; Ochoa, Jose M et al. (2017) Sensor Fusion to Infer Locations of Standing and Reaching Within the Home in Incomplete Spinal Cord Injury. Am J Phys Med Rehabil 96:S128-S134
Shawen, Nicholas; Lonini, Luca; Mummidisetty, Chaithanya Krishna et al. (2017) Fall Detection in Individuals With Lower Limb Amputations Using Mobile Phones: Machine Learning Enhances Robustness for Real-World Applications. JMIR Mhealth Uhealth 5:e151