Êtes-vous un étudiant de l'EPFL à la recherche d'un projet de semestre?
Travaillez avec nous sur des projets en science des données et en visualisation, et déployez votre projet sous forme d'application sur Graph Search.
This thesis introduces spectroscopy-free Raman biosensing, which may find increasing use in the next generation of wearable devices for preventive healthcare. While wearables have made substantial advancements in detecting physical biomarkers, they have yet to effectively evaluate the biological and chemical biomarkers associated with body functions. The potential to monitor the patterns or alterations in chemical biomarkers over time presents a promising avenue for identifying disease precursors before clinical symptoms manifest, offering a window for early intervention. However, the dependence of the existing biosensors on biorecognition elements derived from natural sources, such as animals and plants, precludes a "wear and forget" operation in wearable devices. These bioreceptors, with finite lifespans outside their natural environment, progressively deteriorate over time, mandating the periodic replacement of biosensors. In contrast, the Raman effect offers a unique capability to translate the specific vibration modes of molecules into distinctive fingerprint spectra, suggesting a promising biosensing transducer without the reliance on bioreceptors. However, the instrumentation of Raman spectroscopy remains bulky and unsuitable for wearable applications. Consequently, this thesis demonstrates departure from conventional spectroscopic applications by showcasing the efficacy of Raman shift detection for non-invasive biofluid analysis, which obviates the necessity for one of the most oversized components in Raman instrumentation. First, in static measurements, we demonstrate highly sensitive and selective Raman biosensing of sweat lactate and urea through their single characteristic Raman bands. Subsequently, we engineer a soft epidermal microfluidic device and optimize its integration with Raman biosensing to facilitate in situ monitoring. Our development cycle involves systematic material selection and numerical simulations to optimize the kinetics of microfluidics for effective sweat collection, transportation, and subsequent evaporation post-sampling. Our two-step facile fabrication process utilizes vertically stacked laser-patterned soft materials. We conduct comprehensive investigations with microfluidic-assisted Raman biosensing to validate the device under stationary and dynamic fluid flow conditions. Our findings demonstrate that the microfluidics stabilizes the background of Raman spectra, negating the necessity for its removal. Finally, this thesis introduces spectroscopy-free Raman shift detection achieved by integrating a low-light CMOS image sensor coupled with an optical narrow bandpass filter into a compact optical system crafted solely from off-the-shelf optics and 3D-printed mounts. The compact dimensions of our prototype enable mounting the system on the arm for sedentary sweat analysis post-collection. Extensive testing confirms the alignment precision and optical functionality of the system surpassing those of a conventional bench-top instrument by minimizing optical loss experienced by backscattered Raman photons. A comprehensive measurement shows selectivity (over +30 sweat analytes), stability (across 90 days), and accuracy (assessed against standard methods) of Raman biosensing in ex vivo settings. Therefore, our initial findings catalyze continued exploration, given the potential of the proposed solutions to tackle prevalent challenges, thereby paving the way for advancements in miniaturizing Raman systems.
Hatice Altug, Aleksandrs Leitis, Ming-Lun Tseng
Majed Chergui, Andre Al Haddad, Lars-Hendrik Mewes, Rebecca Ann Ingle