Multi-modal uncertainty analysis of physiological measurements from wearable devices, for clinical applications

About the project

About the Project This project aims to address the critical gap between the huge potential offered by wearable sensors and their still very limited adoption in clinical practice. This will be achieved by creating scientifically rigorous data processing pipelines that provide users with clear and interpretable information about data accuracy and uncertainty. This project builds on an 8-year research collaboration between the Primary Supervisor (Dr Ferrero) and the Partner (Dr Kumar, a consultant paediatric neurologist), aimed at increasing the clinical adoption of wearable technology, through the development of scientifically rigorous software tools that provide clinicians with clear and interpretable information about the accuracy and reliability of data collected by wearable sensors (photoplethysmography, accelerometry, electrodermal activity, and others). The uniqueness of this team’s research direction, compared to other research carried out in this area, is the focus on a scientifically rigorous and interpretable uncertainty quantification of wearable data, which is missing from the vast majority of studies, especially those using machine learning techniques. An accurate uncertainty quantification is of critical importance for two main reasons: 1) to allow correctly fusing information from multiple sensors, and 2) to allow clinicians or other users to make informed decisions based on the measured data, by distinguishing between actual changes in relevant physiological indicators and artifacts caused by noise/disturbance or signal processing errors. The Primary Supervisor is uniquely positioned to lead this project, as he is a globally recognised expert in electrical/electronic instrumentation and measurement and uncertainty analysis, as confirmed by his Editor-in-Chief role for the leading international journal in this field, the IEEE Transactions on Instrumentation and Measurement. Some of the most notable achievements of this research group to date are: 1) A world-first initial attempt to quantify the uncertainty of heart-rate-related indicators calculated from photoplethysmography (PPG) data corrupted by artifacts, through rigorous mathematical analysis (journal paper currently under review). 2) The novel and potentially ground-breaking application of a time-and-frequency-domain method (based on the Taylor-Fourier transform) for the extraction of heart rate and heart rate variability from PPG signals, providing a much better time resolution compared to the state of the art, while also assessing the signal quality [1-3]. In this project, the candidate will build on the existing research, with the following key objectives: 1) Expand the work done on PPG signals, considering also other commonly measured signals (accelerometry, electrodermal activity, and possibly others), to improve the measurement accuracy and uncertainty quantification by fusing information from all those signals. 2) Combine the already-developed Taylor-Fourier analysis with other tools suited for uncertainty quantification (e.g. Gaussian processes) and possibly machine learning techniques, if appropriate. 3) Implement the developed algorithms in a device-agnostic software, allowing clinicians to seamlessly analyse and combine data recorded from a variety of different wearable devices. The student will be based at the University of Liverpool but will work closely with the Partner (also based in Liverpool) throughout the duration of the project. Opportunities for other collaborations may also be available. Interested applicants who want to know more about the project are warmly invited to contact the primary supervisor, Dr Roberto Ferrero ( Roberto.Ferrero@liverpool.ac.uk ). Dr Roberto Ferrero (Primary Supervisor) Roberto.Ferrero@liverpool.ac.uk Dr Alessandro Varsi (Secondary Supervisor) A.Varsi@liverpool.ac.uk Dr Ram Kumar (External Supervisor) ram@kles.co.uk Candidates wishing to apply should complete the University of Liverpool application form to apply for a PhD in Electrical Engineering and Electronics. Please review our guide on How to apply for a PhD | Postgraduate research | University of Liverpool carefully and complete the online postgraduate research application form to apply for this PhD project. Please ensure you include the project title and reference number ENGDLA001 when applying. This UKRI funded Studentship will cover full tuition fees (for 2025-26 this is £5,006 pa.) and pay a maintenance grant for 3.5 years, at the UKRI standard rates (for 2025-26 this is £20,780 pa.) The Studentship also comes with access to additional funding in the form of a Research Training Support Grant to fund consumables, conference attendance, etc. We want all of our Staff and Students to feel that Liverpool is an inclusive and welcoming environment that actively celebrates and encourages diversity. We are committed to working with students to make all reasonable project adaptations including supporting those with caring responsibilities, disabilities or other personal circumstances. For example, If you have a disability you may be entitled to a Disabled Students Allowance on top of your studentship to help cover the costs of any additional support that a person studying for a doctorate might need as a result. We believe everyone deserves an excellent education and encourage students from all backgrounds and personal circumstances to apply. References: [1] S. Rahbar et al., IEEE Trans. on Instrum. and Meas. , 2025, http://www.doi.org/10.1109/TIM.2025.3621748 . [2] S. Rahbar et al., 2024 IEEE I2MTC , Glasgow, UK, http://www.doi.org/10.1109/I2MTC60896.2024.10560902 . [3] S. Rahbar et al., 2025 IEEE I2MTC , Chemnitz, Germany, http://www.doi.org/10.1109/I2MTC62753.2025.11079169 .

Research areas