Optical diagnostics for oral and gastrointestinal tract pathologies

Oral potentially malignant disorders (OPMD) are a significant global health burden. They are a precursor to oral cancer and indicate a tissue with a higher risk of developing a malignancy. Early diagnosis can improve patient outcomes by minimally invasive treatment, reduced mortality rate and improved post-treatment life quality. Despite the availability of advanced technology and diagnostic tools, early detection of OPMD is limited by reliance on invasive biopsies and histopathology, which are time consuming and uncomfortable for patients, and are not suitable for widespread screening or routine monitoring in a high-risk population. Moreover, this diagnostic method is prone to inter-observer and intra-observer inconsistencies and sampling error. Although technological advancements have improved diagnostic capabilities, limitations remain in accessibility, cost-effectiveness, and sensitivity; particularly for low-resource settings. Emerging diagnostic methods, including imaging, molecular assays and optical techniques, offer promise but lack comprehensive validation and seamless integration into clinical workflow. Most available technologies fail to address critical gaps such as non-invasive early detection, real-time monitoring, and the ability to link findings to actionable clinical outcomes. Furthermore, despite the biological relevance of saliva in oral health diagnostics, its potential remains underutilized due to challenges in standardization and correlation with tissue-specific markers. This project is an exploration of what non-invasive diagnostics can achieve, when innovative methodologies meet a persistent clinical challenge. At its core, this work reflects an attempt to democratize OPMD diagnostics, shifting narrative from invasive, resource-intensive and complex procedure to non-invasive, real-time and accessible solution. This thesis explores the development and validation of a multimodal spectroscopic approach and its application in OPMD diagnosis and screening. This multimodal approach was developed to enable in vivo tissue and ex vivo saliva analysis in the clinical setting. The developed system incorporates customized modalities of Raman and diffuse reflectance spectroscopy and proposes a multi-tiered diagnostic approach to enhance clinical decision-making. The clinical study involved the collection and analysis of saliva and in vivo tissue analysis from OPMD patients and healthy controls. Saliva screening was investigated as a non-invasive first-line diagnostic tool to detect biochemical alterations associated with OPMD, using Raman spectroscopy of dried and liquid samples. For individuals flagged through saliva screening, in vivo tissue analysis is proposed using diffuse reflectance spectroscopy (DRS) and Raman spectroscopy to offer functional insights into morphology, oxygenation and chemical changes in the tissue. Machine learning models, including Lasso regression and Linear Discriminant Analysis (LDA), were employed to identify key spectral features and classify samples with high sensitivity and specificity. This multimodal approach provides complementary information. DRS probes optical properties such as absorption and scattering that helps in depicting chromophore’s (including Hb, water, lipid) concentration, oxygenation and tissue structure. Raman spectroscopy, in contrast, provides more specific biochemical information regarding conformational and compositional changes in proteins, lipids, nucleic acids and related metabolites. Although some overlap exists, their combination offers extended diagnostic potential by linking functional and molecular alterations within the same tissue region, thereby improving the diagnostic potential and risk stratification. Building on this progression, Raman analysis of liquid saliva was advanced using photonic crystal fibers (PCFs), which amplify Raman signals to detect subtle biochemical markers in real-time. This enabled label-free, point-of-care screening that complements tissue diagnostics. Together, these techniques create a clinically relevant framework: saliva screening through dry saliva sample and PCF-based Raman spectroscopy acts as a gatekeeper for regular monitoring; and in vivo tissue evaluation, provides a pathway for biopsy guidance, intraoperative decision making and follow-up, addressing the challenges of sampling errors and incomplete resection. The results of this thesis highlight the feasibility of integrating Raman and DRS into clinical workflows to complements traditional biopsy. By addressing diagnostic gaps with a focus on clinical utility, this work establishes a robust, scalable, and non-invasive approach to improve the OPMD screening, early detection, monitoring, and management of OPMD. This work provides a strong foundation for future studies aimed at refining these techniques and expanding their clinical applicability. By bridging engineering innovation with clinical needs, this research advances the goal of making early OPMD diagnostics more accessible, accurate, and patient-friendly.

Grantor dc:publisher

University College Cork

Year dc:date.issued

2025

Author dc:creator

• Maryam, Siddra

Advisors dc:contributor.advisor

• Andersson-Engels, Stefan
• Burke, Ray
• Ni Riordain, Richeal

• Optical diagnostics
• Oral potentially malignant disorders
• Oral cancer
• Saliva screening
• Optofluidics
• Diffuse reflectance spectroscopy
• Raman spectroscopy

Statement dc:rights

• © 2025, Siddra Maryam.

Licence dc:rights.uri

• https://creativecommons.org/licenses/by/4.0/

Language dc:language.iso

en

Handle dc:identifier.uri

https://hdl.handle.net/10468/18447

OAI identifier oai:identifier

oai:cora.ucc.ie:10468/18447