Rapid Detection of Glucose on Nanostructured Gold Film Biosensor by Surface-Enhanced Raman Spectroscopy

SERSNet: Surface-Enhanced Raman Spectroscopy Based Biomolecule Detection Using Deep Neural Network

Surface-Enhanced Raman Spectroscopy (SERS)-based biomolecule detection has been a challenge due to large variations in signal intensity, spectral profile, and nonlinearity. Recent advances in machine learning offer great opportunities to address these issues. However, well-documented procedures for model development and evaluation, as well as benchmark datasets, are lacking. Towards this end, we provide the SERS spectral benchmark dataset of Rhodamine 6G (R6G) for a molecule detection task and evaluate the classification performance of several machine learning models. We also perform a comparative study to find the best combination between the preprocessing methods and the machine learning models. Our best model, coined as the SERSNet, robustly identifies R6G molecule with excellent independent test performance. In particular, SERSNet shows 95.9% balanced accuracy for the cross-batch testing task.

Quantitative, Temperature-Calibrated and Real-Time Glucose Biosensor Based on Symmetrical-Meandering-Type Resistor and Intertwined Capacitor Structure

Here, we propose a glucose biosensor with the advantages of quantification, excellent linearity, temperature-calibration function, and real-time detection based on a resistor and capacitor, in which the resistor works as a temperature sensor and the capacitor works as a biosensor. The resistor has a symmetrical meandering type structure that increases the contact area, leading to variations in resistance and effective temperature monitoring of a glucose solution. The capacitor is designed with an intertwined structure that fully contacts the glucose solution, so that capacitance is sensitively varied, and high sensitivity monitoring can be realized. Moreover, a polydimethylsiloxane microfluidic channel is applied to achieve a fixed shape, a fixed point, and quantitative measurements, which can eliminate influences caused by fluidity, shape, and thickness of the glucose sample. The glucose solution in a temperature range of 25-100 °C is measured with variations of 0.2716 Ω/°C and a linearity response of 0.9993, ensuring that the capacitor sensor can have reference temperature information before detecting the glucose concentration, achieving the purpose of temperature calibration. The proposed capacitor-based biosensor demonstrates sensitivities of 0.413 nF/mg·dL-1, 0.048 nF/mg·dL-1, and 0.011 pF/mg·dL-1; linearity responses of 0.96039, 0.91547, and 0.97835; and response times less than 1 second, respectively, at DC, 1 kHz, and 1 MHz for a glucose solution with a concentration range of 25-1000 mg/dL.