Twitting falls among elderly people

NTNU students in collaboration with SINTEF ICT Health Informatics Group use social media to detect and manage falls among elderly people.  The solution allows family and neighbors to participate in fall detection and dispatching help.

Group 11 of IT2901 at IDI, NTNU, from left to right: Silje Grimstad Vikås, Simen Kjellgren, Trond Klungervik, Ada A. Jordal, Henrik Linde og Eivind Lysne

Physical falls and consequent injuries among elderly people are more common and more damaging than what most of us can imagine. As we get older, our vision and hearing get impaired, our bones get weaker, we take more medication, and as a result we become prone to physical falls. Physical falls cost enormously, both physically, emotionally, and economically. Once a fall occurs, chances for prolonged hospitalization and further falls increase. Falls also result in the deterioration of the social lives of those who are affected.

You might think that Facebook is only for fun and leisure, for sharing photos and talking about music. But potentially leisure is only one very early application of social media. In fact, Group 11 in IT2901 at NTNU/IDI has used features from social media to implement a fall detection solution for elderly. One of the major problems we face in fall management is early detection of falls. A person who has fallen is often in severe pain and probably in shock. To know early that someone has fallen and to dispatch help is crucial for the involved parties, both the injured person but also his/her family. The idea of using one’s social network in early fall detection and help dispatching came from Arild Faxvaag of NSEP (Norsk Senter for Elektronisk Pasientjournal). It was obvious from the beginning that this was a challenging task, both technically and organizationally. Group 11 used the persona method to do a stakeholder analysis and worked on requirements, architecture and design of the system based on this early analysis. The developed solution provides a basis for a real-life trial of social fall detection in Trondheim Norway, as part of the European project FARSEEING.

About the solution

The user interface provided to health care personnel allows them to manage groups and monitor falls.

The social fall detection prototype developed by Group 11 of NTNU/IDI IT2901 provides the following features:

  • A user interface for health care personnel (e.g. at a hospital or an elderly care center) to manage groups of people who participate in fall detection of a particular elderly person. This group can include family members, neighbors or close friends of the elderly person.
  • A user interface for the on-duty health care personnel to monitor falls (see the Figure to the left).
  • A user interface for the elderly person (on an Android phone) to allow communication with the group monitoring him/her.
  • A user interface for the group members (family, neighbors, friends) to be notified of falls, rescue status, and possible requests for help.

Group 11 was asked to use some cutting edge technologies (e.g. Shinding, XMPP) to implement the solution. In this perspective this task was as much a research project as it was a development project. The innovative service model behind the solution will be evaluated in real world settings later on during 2013 in Trondheim area.

About course IT2901

The course IT2901 at IDI, NTNU is a practical programming project course that exposes students to real world programming problems. The course has invited customers from the industry or other external institutions. Professor Monica Divitini is the coordinator of the course. We acknowledge her support in making this project possible.