Optical and Wearable Detection of Diabetes

Abstract

This thesis presents a thorough investigation into detection of diabetes, incorporating cutting-edge optical technologies alongside advanced machine learning algorithms. The emphasis centers is on the utilisation of Surface Plasmonic Resonance (SPR) and Localized Surface Plasmonic Resonance (LSPR) technologies, which have been creatively utilised to devise precise and selective approaches for the detection of diabetes biomarkers, including Glycated Albumin (GA).

This study explores the potential of LSPR technology for the detection of GA, a crucial biomarker for diabetes. The research focuses on the utilisation of cost-efficiently manufactured gold nanoparticles. This methodology offers a more precise and cost-effective alternative for the detection of diabetes in contrast to conventional approaches.

Furthermore, a substantial portion of the study shows the advancement and execution of machine learning models, including deep learning methodologies such as Convolutional Neural Networks (CNNs), to process intricate medical data. These models are employed for the processing and interpretation of data derived from many sources, including electrocardiogram (ECG) signals. This utilisation provides a new and innovative approach in the timely identification and treatment of diabetes. Using the physiological signal, this thesis explores the domain of wearable technologies, specifically examining the use of complex ECG signal for diabetes detection. The objective of this strategy is to improve patient adherence and convenience, hence progressing towards a healthcare management system that is more tailored and proactive.

This thesis critically examines the limitations, problems, and future directions of these technologies in clinical contexts. The research reported in this study not only makes a valuable contribution to the progress of non-invasive approaches for detecting diabetes, but also establishes a foundation for potential future advancements in the field of medical diagnostics and wearable healthcare technologies. Show more