Data captured by body worn sensors provides an excellent opportunity for assessing the physical activity of patients and hence creating behavioral profiles over time. Particularly patients with chronic disease can receive tailored advice on how to increase their activity and hence improve their overall life quality.
The focus and challenge for this project and master thesis is the selection, implementation and improvement of machine learning to identify different types of activities such as sitting, standing, walking, running, biking, etc within sensor data.
Over the last years we have collected and annotate several datasets which will be provided for the thesis. Moreover as this work is related to the HUNT4 study (https://www.ntnu.no/hunt4/), that collected recordings from over 30 000 participants.
During this work, the state of the art for algorithms should be described, the most fitting algorithms selected and implemented. We have already implemented different approaches which can be improved in terms of increasing accuracy for a larger variety of participants or new activities can be added. In this work we would like to focus on high intensity activities that have a significant effect on the health of people.
Once the experimental set up is created, different existing algorithms should be evaluated and their strength and weaknesses pointed out. Based on this analysis, a follow-up master thesis can be defined focusing on improving existing algorithms and validated in a real world setting.
Sketch for the project thesis
Sketch for a follow-up master thesis
For more information on this thesis, feel free to contact Kerstin Bach (email@example.com). We will then set the specific topic and scope of your thesis together.