About the Project
Smartphones are an essential part of everyday life for many people and commodity wearables, including smart watches, are now widely available. More recently, earables -- smart earphones with advanced sensing capabilities -- have emerged as the latest commodity wearable to attract research interest . Thanks to advances in sensing technologies these devices come equipped with a wide range of sensors, including inertial measurement units (IMU), and AI-based technologies are now being used to unlock novel and unique applications by combining sensor readings from phones, watches, and earables simultaneously .
This project will explore how AI technology can be used to monitor spinal mobility of patients that suffer from Ankylosing Spondylitis (AxSpa) using data from commodity wearables. AxSpa is a form of Spondyloarthritis affecting as much as 0.3% of the global population. The disease primarily affects the joints in the hips, back, and neck causing stiffness, chronic pain, restriction of mobility, and eventually fusion of the spine. In addition, acute worsening of symptoms, known as ‘flare ups’, can occur which affects the patient’s quality of life [3, 4, 5]. The Bath Ankylosing Spondylitis Metrology Index (BASMI) is currently used to measure changes in spinal movement due to its simplicity, however the index itself lacks sensitivity and specificity . In addition, the BASMI, measured by trained physiotherapists or clinicians, is not being performed as regularly as possible due to the limited availability of resources.
In the first part of this project, we will investigate how AI-based algorithms can learn effective mappings to measure spinal mobility from commodity sensors, combined with easy-to-wear bespoke sensors (such as a belt-based device). Initial research has shown how protocols using IMUs have the potential to act as a novel mobility index using controlled protocols involving BASMI-like functional movements to elicit the required motion . We will relax these constraints to explore the feasibility of patients being able to assess their own mobility at home using commodity wearables and smart devices. In the second part of the project, the aim is to gather data from patients wearing basic commodity wearables as they go about their everyday lives. These data streams will be combined using an AI-based system that aims to monitor the mobility of the patient autonomously, while detecting potential flare-ups. The ability to monitor progression of AxSpa discreetly and unobtrusively while automatically detecting flare ups has the potential to better inform treatment regimens, transforming patient care and quality of life. The system can provide both the patients and health professionals with a more detailed picture of spinal mobility and its variation over time.
Despite the potential advantages of the proposed system, transparency and accountability are paramount due to potential use in a clinical setting. The validity of the proposed approach will be assessed using clinical measures and the project will be supported by the lead Consultant for ankylosing spondylitis patients at the Royal National Hospital for Rheumatic Diseases in Bath. It is paramount that both clinicians and patients are in-the-loop during development and evaluation phases. The project also raises questions about how best to maximise transparency around the limitations of the system and the datasets used to train the underlying models, while algorithm and model selection should be driven with rheumatologists to ensure system outputs are interpretable and accurate enough for them to make informed decisions .
This project is associated with the UKRI Centre for Doctoral Training (CDT) in Accountable, Responsible and Transparent AI (ART-AI). We value people from different life experiences with a passion for research. The CDT's mission is to graduate diverse specialists with perspectives who can go out in the world and make a difference.
Informal enquiries about the project should be directed to Dr Clarke.
Applicants should hold, or expect to receive, a First or Upper Second Class Honours degree or a Master's degree in a relevant subject. You will also need to have taken a mathematics course or a quantitative methods course at university or have at least grade B in A level maths or international equivalent. Programming experience is desirable.
Formal applications should be accompanied by a research proposal and made via the University of Bath’s online application form. Enquiries about the application process should be sent to email@example.com.
Start date: 2 October 2023.
 Röddiger, T., Clarke, C., Breitling, P., Schneegans, T., Zhao, H., Gellersen, H., & Beigl, M. (2022). Sensing with Earables: A Systematic Literature Review and Taxonomy of Phenomena. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 6(3), 1-57.
 Strömbäck, D., Huang, S., & Radu, V. (2020). Mm-fit: Multimodal deep learning for automatic exercise logging across sensing devices. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 4(4), 1-22.
 Rafia, R., Ara, R., Packham, J., Haywood, K., & Healey, E. (2012). Healthcare costs and productivity losses directly attributable to ankylosing spondylitis. Clinical and Experimental Rheumatology-Incl Supplements, 30(2), 246.
 Dagfinrud, H., Kjeken, I., Mowinckel, P., Hagen, K. B., & Kvien, T. K. (2005). Impact of functional impairment in ankylosing spondylitis: impairment, activity limitation, and participation restrictions. The Journal of rheumatology, 32(3), 516-523.
 Krüger, K., von Hinüber, U., Meier, F., Tian, H., Böhm, K., Jugl, S. M., ... & Braun, S. (2018). Ankylosing spondylitis causes high burden to patients and the healthcare system: results from a German claims database analysis. Rheumatology international, 38(11), 2121-2131.
 Martindale, J. H., Sutton, C. J., & Goodacre, L. (2012). An exploration of the inter-and intra-rater reliability of the Bath Ankylosing Spondylitis Metrology Index. Clinical rheumatology, 31(11), 1627-1631.
 Franco, L., Sengupta, R., Wade, L., & Cazzola, D. (2021). A novel IMU-based clinical assessment protocol for Axial Spondyloarthritis: A protocol validation study. PeerJ, 9, e10623.
 Giordano, C., Brennan, M., Mohamed, B., Rashidi, P., Modave, F., & Tighe, P. (2021). Accessing artificial intelligence for clinical decision-making. Frontiers in Digital Health, 3, 65.