Fall Detection using Mobile devices
Have you ever fallen down, while walking or sleeping? Do you know people who tend to fall very often? Do you usually trip over things? I do... I do know people who tend to fall very often.
Falling is an everyday potential accident that all of us are exposed to. A fall can cause injuries or hurt people. Falls can to a large extent be prevented though. There is even an own field of research devoted for fall prevention. But not all falls can be prevented, especially when there is ice on the road or it is slippery. At least we can try to detect fall accidents to provide quick assistance if required.
Based on this idea, master student in Media Technology , Ian Gonzales is writing his master thesis on fall detection using mobile phones. The main goal of his project is to create an application for mobile devices that will detect when a person falls down. The application, with the aid of the mobile's orientation sensors (accelerometers, etc.) and smart analysis, will determine if the person needs assistance. In case that a person needs any kind of assistance or help, the application will send a message to pre-configured health services such as an ambulance or a contact person to provide quick assistance.
In 2009, Norway's population was 4,827,038. 15% of the total population is over 65 years old, an age group which tend to fall more often than others. The main target users for this application will of course be elderly people, but the application is not exclusive for their use. It doesn't matter if you are young or old, tall or small, falling can happen to you, or even me. It might also be useful for adventurous people who likes to go hiking alone and are exposed to suffer an accident without anyone near to help them.
Detecting a fall with a mobile device is not as easy as it seems. It can be misinterpreted when a person jumps or simply lies down on a bed. Therefore a lot of research and experiments have to be done. Gonzales has stated that gathering data from an actual person is a difficult part of his research. The main reasons are that it is time consuming and that it moreover is not easy to convince a person to participate when you ask them to fall down, possibly getting hurt in the process. The first part of the experiment consists in simulating different activities, mainly Activities for Daily Living (ADL), which are activities that you normally perform in a day-to-day basis, such as standing or walking. These activities are measured by a mobile phone that the participant needs to carry while performing the experiment. After gathering and analyzing this data, a second session needs to be done to analyze actual falls. Because this testing may be painful for the participant, some testing will be performed using a dummy, facilitated by the nursing department at HiG.
Gonzales expects his master thesis to be of great help to society, mainly for the health community, and especially for elderly people in preventing serious physical damage to a lot of people and providing them quick assistance when required.
This research is what HiG students do to make the difference!
Ian (email@example.com) has been a Media Technology master student since August 2009. He has a bachelor degree in Information in Technology from University of Oslo.