Continuous ambulatory blood pressure monitoring using optical fibre sensors
Publication Date
December 9, 2025
Creator
Abstract
Abnormal blood pressure (BP) is a key indicator of cardiovascular dysfunction and a leading risk factor for stroke, heart failure, and mortality. Hypertension affects more than a quarter of adults in England, costing the NHS approximately £2.1 billion annually. Despite effective therapies, rates of uncontrolled BP remain high, underscoring the need for earlier detection and continuous monitoring. Beat-to-beat BP measurement offers distinct advantages by capturing rapid cardiovascular dynamics, yet current methods face limitations: invasive catheters are unsuitable for long-term use, oscillometric cuffs lack temporal resolution, and photoplethysmography (PPG) is prone to artefacts, latency, and skin-tone variability.
This thesis presents a novel, non-invasive cuffless blood pressure (BP) monitoring system that employs fibre Bragg grating (FBG) cantilever sensors to estimate pulse transit time (PTT) and integrates electrocardiography (ECG) to derive pulse arrival time (PAT). The proposed system offers a light-insensitive and motion-resilient solution capable of high-fidelity pulse detection, making it suitable for real-time monitoring across clinical, home, and telemedicine settings.
A validated pipeline -- spanning mechanics, temperature-strain decoupling, phantom haemodynamics, and in-human exercise trials -- demonstrates technical feasibility, with a biocompatible FBG cantilever and compact interrogator achieving tens-of-milliseconds timing resolution. In human trials, PAT showed strong inverse correlations with systolic BP and heart rate, supporting its use in cuffless monitoring. Algorithmically, a GAN-based reconstruction stage restored morphology under artefacts, while Bayesian Gaussian Process Regression (GPR) produced calibrated mmHg estimates (systolic ≤10 mmHg, diastolic ≤4 mmHg) and modelled inter- and intra-subject variability.
By combining high-sensitivity optical sensing, probabilistic modelling, and AI-based signal reconstruction, this work establishes a new pathway toward continuous, personalised BP monitoring. The findings have direct implications for next-generation medical devices that support early diagnosis, community screening, and digital health strategies aimed at reducing the global burden of hypertension.
Item Type
ethesis
Thesis Type
PhD
Subjects (LC)
Associated Schools / Departments
Department of Electrical and Electronic Engineering (UK)
eprints ID
82318
UoN Repository URI
Except where otherwise noted, this item's license is described as
File(s)![Thumbnail Image]()
Name
Thesis_corrected_12Sep25.pdf
Type
Full-text
Description
Examined
Size
119.42 MB
Format
Adobe PDF
Checksum (MD5)
41526b7f562687014fccc3e0a90f9493