Advanced health behaviour assessment in free living
The projects intends to integrate advanced behaviour inference algorithms for sleep, diet, and other daily rhythms based on various smartphone sensors to investigate health trends in patients outside the hospital. The work will start from existing pattern inference algorithms and extend them with dynamic model learning and personalisation. The project can utilise a cloud-based processing framework to evaluate algorithms and manage smartphone data. In collaboration with Prof. Zopf, Hector Center, UK Erlangen data from a patient cohort will be analysed regarding risk factors related to diet and exercise. Research in this project will focus on attribute-weighted deep learning methods and expert models. The project has potential to create important new insight on treatment options and remote patient diagnosis measured daily behaviour and thus can provide an ideal basis for a scientific publication and research towards the Ph.D.
|Project type||Master Thesis|
|Work distribution||30% experimental, 70% algorithm development|
|Useful knowledge||Machine learning, Android/IOS programming, signal processing|