Early diagnosis of movement-related diseases and risk of falls using wearable sensors

Falls in ageing population is a global challenge. One in three people over 65 fall each year, rising to one in two for adults over 80. Recurrent falls are associated with increased mortality, increased hospitalization and higher rates of long-term care. Early diagnosis of deterioration of balance and increased risk of fall can provide the care system with an opportunity to avoid negative consequences by providing timely treatment. A large amount of physiological (e.g. gait data) and biological (e.g. blood pressure and heart rate) data can potentially be measured using wearable sensors. However, the potentials of these big data have not been fully exploited.
The recent advances in signal processing and artificial intelligence have made it feasible for the computer to be trained to aid the diagnosis and prediction of various diseases. This project aims to develop new methodologies in the field of signal processing and artificial intelligence (AI) (e.g. deep learning) that support automated and accurate diagnosis and prediction of movement-related disease and risk of falls. The wearable sensory system developed in the Human-Environment Dynamics (HED) Laboratory at the University of Bath will be used to collect a uniquely comprehensive and detailed physiological and biological dataset from aging population outside laboratory environment. The collected data then will be used to develop a computational model for early and accurate diagnosis of risk of falls (using signal processing and AI methods). The real-world data will be benchmarked with detailed laboratory measurement using the extensive gait analysis equipment available in the HED laboratory. The focus of the project will initially be on the estimation of ‘risk of falls’, but if time allows, the collected data can also be analysed for other movement related diseases.
The PhD candidate will be able to use the state-of-the-art experimental and technical facilities of both the HED laboratory of the University of Bath and the Vibration Engineering Section (VES) at University of Exeter including visual and LIDAR-based movement monitoring systems, optical motion captures, force plates, instrumented treadmills, wearable inertial measurement units, and EMG, EEG and ECG monitoring systems, etc.
The PhD candidate ideally (but not essential) will be famililar with gait analysis, gait/physiological signal processing, computational modelling particularly neural networks and pattern recognition and a programming language such as Matlab or Pyton.