| dc.description.abstract | Optical biosensors are devices that employ light to detect the presence or concentration of a biological analyte. These devices play an important role in modern society in areas that include medical diagnostics, infectious disease screening, food safety and environmental monitoring. Porous silicon (PSi), a material that features large specific surface area, versatile surface chemistry, tuneable optical properties and simple fabrication, has been widely studied as a promising biosensing platform. In this thesis, we explore the use of PSi-based label-free optical biosensors towards point-of-care diagnostics (POCD) applications, with the goal of overcoming challenges that often prevent benchtop lab instruments that require trained personnel from being adapted to simple biosensors that can be operated in a straightforward manner outside the laboratory. One major challenge for POCD sensors is the cost, which limits their availability for widespread distribution and use. To address this issue, we demonstrate that by using PSi as the transducer, along with a smartphone LED as the light source and the embedded camera in the smartphone as the photodetector, a stand-alone biosensing platform can be realized for low-cost biosensing. A second challenge for POCD sensors is the instability of some bioreceptors in harsh conditions that are more common in low resource environments. To improve the environmental stability of PSi biosensors, we demonstrate that peptide capture agents can replace less stable protein antibody bioreceptors. A third challenge for POCD sensors is identifying constituent species in a test sample due to the need to include a different bioreceptor for each of the possible constituent species. Inspired by the optoelectronic nose for gas and chemical sensing, we propose a biorecognition element-free biosensing system based on a PSi array with individual PSi array elements exposed to a variety of different physiochemical conditions. The proposed system utilizes machine learning to reveal the link between sensor data from the array and the properties of the target biomolecules. Overall, the advances in integrating smartphones, peptides, and machine learning with an already promising PSi biosensor platform represent significant progress towards realizing a cost-effective, robust, and versatile point-of-care biosensor. | |
