Our partners submitted a paper at the IEEE International Conference on E-health Networking, Application & Services that took place on 17-20 September 2018 in Ostrava, Czech Republic. We are glad to announce that the paper received the 2nd best paper award. We congratulate our team for this wonderful news.
Executive summary of the paper “Meeting challenges of activity recognition for ageing population in real life settings”:
As the global community becomes more interested in improving the quality of life of older people and preventing undesired events related to their health status, the development of sophisticated devices and analysis algorithms for monitoring everyday activities is necessary more than ever. Sophisticated methods for the detection of distinct physical activities have been reported in the literature, using a variety of wearable and non-wearable sensors. However most of the frameworks for behavioral monitoring have been tested on data from young and healthy participants, or the experiments were performed on laboratory conditions or using simulated data. Those works report high classification accuracy, but results are not directly comparable with uncontrolled monitoring systems in real home environments.
Creating robust models from data collected unobtrusively in home environments can be challenging, especially for the vulnerable ageing population. Under that premise, we propose an activity recognition scheme (discriminating the activities like sit/stand, laying, walking, walking upstairs/downstairs, transition between activities) for older people along with heuristic computational solutions to address the challenges due to inconsistent measurements in non-standardized environments. In more details, moving from laboratory environment to real-life experiments, researchers are dealing with numerous obstacles that they must overcome, concerning mostly the devices used to monitor older people.
Thanks to our researchers’ contributions in order to overcome the observed obstacles, the obtained accuracy was significantly increased compared to the previous (standard) methodology.
The full paper is available here.