Restriction lift date: 2024-05-31
Optical spectroscopy for biological and biomedical applications: potentially impacting future of healthcare with research, clinical translation and education
No Thumbnail Available
Saito Nogueira, Marcelo
University College Cork
Clinical interventions have been significantly improved by recent advances in devices for disease detection, monitoring, diagnosis and surgical guidance. However, most of the current imaging technologies provide primarily morphological/structural information on analyzed subjects or samples, which is insufficient to detect early-stage diseases and identify tissue structures with enough contrast. Optical spectroscopy and imaging can solve this insufficiency by providing molecular-sensitive tools for applications involving screening, diagnosis, monitoring vital signs, treatment planning and guidance, treatment-outcome prediction, and others. These tools provide non-invasive, real-time, cost-effective, and in situ interrogation of biological samples. Key advantages can be added to optical tools upon their miniaturization and integration into existing medical instruments, wearables and portable test kits, as patient prognosis can be improved by increasing the contrast of identification between vital structures and tissues to be resected, as well as by enabling telehealth solutions via remote monitoring individual health status. Such identification and monitoring can be achieved by extracting information on sample microstructure, biochemistry, and associated quantities. The thesis investigates the feasibility of using optical spectroscopy as an optical technique for application in cancer detection and delineation, specifications of next generation optical devices for this application, as well as tissue microstructural and biochemical features associated with carcinogenesis and organ viability for transplantation. Previous studies cover the selection and optimization optical techniques for a limited number of applications and parameters. Most studies of this thesis optimize diffuse reflectance spectroscopy (DRS) parameters for colorectal and organ preservation applications, while instruments developed for other optical techniques and associated analysis methods can be used for a wide range of applications. In particular, the clinical need for CRC diagnosis, surgery and therapy is accurately locating and completely removing premalignant lesions and surrounding compromised tissue. Since clinical procedures have limited time, tools providing endoscopy and surgical guidance must enable real-time detection of CRC and premalignancies. Therefore, we have validated DRS and fluorescence spectroscopy (FS) as optical techniques which could fulfil the clinical need for CRC applications, as well as be integrated into medical devices via fiber optic probes with <4mm source-detector distances (SDD). This validation was performed through pre-clinical and clinical studies and required accuracies and/or area under the receiver operating characteristic curve (AUC) >90% to be used in the clinic. Thesis resources could potentially be used to develop a probe for CRC detection during colonoscopy, laparoscopy, intestinal anastomosis and CRC surgery with the aim of real-time automated tissue classification by using machine learning models coupled with DRS instruments capable of displaying the results of a single reading in 2-3 seconds, and potentially coupled with methods to quantify early-stage cancer biomolecules. To the best of our knowledge, this thesis shows (1) the first DRS study to investigate the potential of probing tissue layers up to 2mm deep by using larger (>600 μm) SDD probes for CRC detection during colonoscopy. (2) the first study evaluating the benefit of the extended DRS wavelength range between 350-1919 nm in colorectal tissues through intra- and inter-study comparison of achieved tissue classification performance metrics, (3) the first study to evaluate biomolecule concentrations and scattering properties of superficial and deeper tissue layers for CRC detection in the luminal wall, (4) the first study evaluating solely the cooling effect on multiple tissue microstructural and biochemical parameters such as blood oxygenation, concentrations of blood, methemoglobin, water, lipid, and bile as well as scattering amplitude, Mie scattering power and fraction of Rayleigh scattering, (5) one of the first studies extracting tissue microstructural and biochemical parameters over extended wavelength ranges between 450-1590 nm by using a spectral fitting based on Monte Carlo simulations of light propagation in complex media, (6) some of the scarce broadband DRS studies investigating tissue measurables such as reflectance and optical properties over wavelengths ranging from 350-1919 nm, and (7) some of the scarce broadband DRS studies combining machine learning methods with algorithms determining tissue microstructural and biochemical parameters.
Biomedical optics , Biophotonics education , Multidisciplinary , Diffuse reflectance spectroscopy , Oral cancer , Colorectal cancer , Near-infrared spectroscopy , Endoscopy , Laparoscopy , Colonoscopy , MedTech , Healthcare , HealthTech , DeepTech , Machine learning , TeleHealth , Intestinal anastomosis , Optical treatments , Phototherapy , Photodynamic therapy , Fluorescence spectroscopy , Fluorescence lifetime spectroscopy , Time-resolved spectroscopy , Time-domain optical spectroscopy , Frequency domain diffuse optical spectroscopy , Optical frequency-domain imaging , Monte Carlo simulations , Light tissue interactions , Optical imaging , Microscopy , Stochastic modelling , Light propagation in complex media , Innovation , Commercialization , Policy making , Gastrointestinal cancer , Medical technology , Digital health , STEM education , Artificial intelligence , Data processing , Multivariate analysis , Statistical analysis , Spatial frequency domain imaging , Tissue optics
Saito Nogueira, M. 2021. Optical spectroscopy for biological and biomedical applications: potentially impacting future of healthcare with research, clinical translation and education. PhD Thesis, University College Cork.