Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/143003
Title: Point-of-care optical biosensor platforms for stroke diagnostics with blood-based biomarkers
Authors: Harpaz, Dorin
Keywords: Engineering::Materials
Engineering::Bioengineering
Issue Date: 2020
Publisher: Nanyang Technological University
Source: Harpaz, D. (2020). Point-of-care optical biosensor platforms for stroke diagnostics with blood-based biomarkers. Doctoral thesis, Nanyang Technological University, Singapore.
Abstract: Stroke, a top leading cause of death, occurs as a result of an acute interruption in the brain blood flow. After stroke onset, brain cells die rapidly. Therefore, therapeutic treatment must be administered early. The ‘time is brain’ rational is a major limitation in therapeutic approaches, due to the association of time delays with worsen outcome in patients. Stroke diagnosis is mainly based on utilizing imaging technologies (magnetic resonance imaging (MRI) or computed tomography (CT)) in order to identify the stroke event in the brain. Stroke is classified to ischemic stroke (blockage) in 87% of cases, or to haemorrhagic stroke (bleeding) in the remaining 13% of cases. Then, ischemic stroke is further classified into different etiologies, based on the stroke mechanism: large artery atherosclerosis (LAA); cardioembolic (CEI); lacunar (LAC); others and undetermined. Stroke etiology classification main value is for therapeutic decision-making process, to allow the admission of time-limited window life-saving therapeutics (IV-tPA), within 4.5 hours from stroke onset. However, ischemic stroke etiologies classification schemes are complexed, time consuming (between hours to days), and require professional personal. Incorporating biomarkers measurement holds big potential as a new stroke etiology classification. Stroke is associated with a variety of pathophysiological changes, which leads to triggering different bio-chemical processes. This results in a big variety of stroke related biomarkers, for which their clinical practice value is yet to be fully determined. An ideal brain biomarker is usually a protein which can be measured from bio-fluids in a safe method, in order to provide additional data on specific organs, mainly the brain and spinal cord. In this thesis, candidate biomarkers were reviewed, and only biomarkers that were available in hospital laboratory analytical platforms were included. The levels of the following multiple biomarkers were investigated: NT-proBNP, D-dimer, S100β, NSE, vitamin D, cortisol, IL-6, insulin, uric acid and albumin. After, the performance of these biomarkers to identify stroke mechanisms was examined, with the use of statistical and data-mining tools. The methodology of using receiver operating characteristic (ROC) analysis was followed in order to identify optimal cut-off values. Multivariable regression was employed using backward stepwise logistic regression model to identify useful biomarkers. Biomarkers measured by a rapid point-of-care biosensor will enable an improved diagnosis approach. A point-of-care test is rapid (results within minutes), user-friendly, robust and can be used on-site. Successful examples are the glucometer and lateral flow pregnancy test. The challenge is to reach the required sensitivity sufficient to make a clinical decision. Several biomarkers were identified as useful to elucidate stroke mechanisms, as well as showed value for additional stroke management needs, such as diagnosis, severity, mortality and therapeutic admission. The use of a multi biomarker panel strategy, instead of the measurement of a single biomarker, is more useful but it is still not used in the clinical practice. After identifying selected and specific stroke biomarkers as potential target analytes, several optical quantitative biosensor platforms technologies were considered and compared. This thesis focuses on developing a point-of-care biosensor platform, for the multiplex and quantitative detection of specific stroke related biomarkers panel, directed specifically to ischemic stroke etiology classification. Four optical biosensor platforms were considered and examined. Two paper-based sensing platforms were examined, based on colorimetric and chemiluminescent signals detection, because of their main advantages for multiplex detection. Moreover, two chip-based (gold-silver and silicon dioxide chips) sensing platforms were examined, based on surface plasmon resonance (SPR) sensing technology, because of their main advantage for sensitive detection. However, each of these four biosensor platforms also exhibit a problematic disadvantage, which must be overcome. First, in order to overcome the disadvantage for sensitive detection of the chemiluminescent and colorimetric paper-based biosensor platform, three different approaches were examined: dissolvable polyvinyl-alcohol film, membrane type comparison and signal amplification. Moreover, in order to overcome the disadvantage for ease of use as point-of-care platform of the SPR biosensor platform, specific functionalized chips from gold, silver and silicon dioxide were developed for use in a miniaturized SPR system. To conclude, the SPR chip-based biosensor platform demonstrated clinically relevant limit-of-detection lower than 1 ng/mL, for both biomarkers tested (NT-proBNP and S100β). While, the paper-based biosensor platforms did not reach the required sensitivity and specificity.
URI: https://hdl.handle.net/10356/143003
DOI: 10.32657/10356/143003
Rights: This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).
Fulltext Permission: embargo_20210717
Fulltext Availability: With Fulltext
Appears in Collections:MSE Theses

Files in This Item:
File Description SizeFormat 
Thesis - Dorin Harpaz - Final - NTU.pdf
  Until 2021-07-17
Thesis - Doctor of Philosophy - Dorin Harpaz - MSE - 20207.65 MBAdobe PDFUnder embargo until Jul 17, 2021

Page view(s)

139
Updated on Apr 21, 2021

Google ScholarTM

Check

Altmetric


Plumx

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.