The development of rapid and accurate biomedical laser spectroscopy systems in the mid-infrared has been enabled by the commercial availability of external-cavity quantum cascade lasers (EC-QCLs). EC-QCLs are a preferable alternative to benchtop instruments such as Fourier transform infrared spectrometers for sensor development as they are small and have high spectral power density. They also allow for the investigation of multiple analytes due to their broad tuneability and through the use of multivariate analysis. This article presents an in vitro investigation with two fiber-coupled measurement setups based on attenuated total reflection spectroscopy and direct transmission spectroscopy for sensing. A pulsed EC-QCL (1200–900 cm ${}^{-1}$ ) was used for measurements of glucose and albumin in aqueous solutions, with lactate and urea as interferents. This analyte composition was chosen as an example of a complex aqueous solution with relevance for biomedical sensors. Glucose concentrations were determined in both setup types with root-mean-square error of cross-validation (RMSECV) of less than 20 mg/dL using partial least-squares (PLS) regression. These results demonstrate accurate analyte measurements, and are promising for further development of fiber-coupled, miniaturised in vivo sensors based on mid-infrared spectroscopy.

Fast and accurate continuous glucose monitoring is needed in future systems for control of blood glucose levels in type 1 diabetes patients. Direct spectroscopic measurement of glucose in the peritoneal cavity is an attractive alternative to conventional electrochemical sensors placed subcutaneously. We demonstrate the feasibility of fast glucose measurements in peritoneal fluid using a fibre-coupled tuneable mid-infrared quantum cascade laser. Mid-infrared spectra (1200–925 cm⁻¹) of peritoneal fluid samples from pigs with physiological glucose levels (32–426 mg/dL, or 1.8–23.7 mmol/L) were acquired with a tuneable quantum cascade laser employing both transmission and attenuated total reflection (ATR) spectroscopy. Using partial least-squares regression, glucose concentrations were predicted with mean absolute percentage errors (MAPEs) of 8.7% and 12.2% in the transmission and ATR configurations, respectively. These results show that highly accurate concentration predictions are possible with mid-infrared spectroscopy of peritoneal fluid, and represent a first step towards a miniaturised optical sensor for intraperitoneal continuous glucose monitoring.

Convolutional neural networks (CNNs) are widely used for image recognition and text analysis, and have been suggested for application on one-dimensional data as a way to reduce the need for pre-processing steps. Pre-processing is an integral part of multivariate analysis, but determination of the optimal pre-processing methods can be time-consuming due to the large number of available methods. In this work, the performance of a CNN was investigated for classification and regression analysis of spectral data. The CNN was compared with various other chemometric methods, including support vector machines (SVMs) for classification and partial least squares regression (PLSR) for regression analysis. The comparisons were made both on raw data, and on data that had gone through pre-processing and/or feature selection methods. The models were used on spectral data acquired with methods based on near-infrared, mid-infrared, and Raman spectroscopy. For the classification datasets the models were evaluated based on the percentage of correctly classified observations, while for regression analysis the models were assessed based on the coefficient of determination (R$^2$). Our results show that CNNs can outperform standard chemometric methods, especially for classification tasks where no pre-processing is used. However, both CNN and the standard chemometric methods see improved performance when proper pre-processing and feature selection methods are used. These results demonstrate some of the capabilities and limitations of CNNs used on one-dimensional data.

Sensors in mid-infrared spectroscopy based on attenuated total reflection (ATR) sensing with internal reflection elements (IREs) facilitate easier measurements of aqueous solutions or other opaque analytes. Micromachined silicon (Si) elements are an attractive alternative to conventional IREs, as they can be produced cheaply with silicon processing. Techniques for surface modifications are also easily integrated into the wafer process, and surface structures such as micropillars or nanoparticles can thereby be used for signal enhancement. Replacing the classic Fourier transform infrared (FTIR) spectrometers with tuneable quantum cascade lasers (QCLs) also opens up new avenues for sensing. In this study, the performance of basic and signal-enhanced Si IREs has been compared for measurements in a spectroscopy setup with a fibre-coupled tuneable QCL source. These IREs had V-shaped microgrooves etched on the underside for more efficient in-coupling of light, while the signal enhanced IREs also had micropillars on the top surface. The results are also contrasted with measurements done in a standard ATR-FTIR spectrometer, using an Alpha II spectrometer with a single-reflection diamond ATR crystal. Various concentrations of glucose (0-5000 mg/dl) in aqueous solutions were used to characterise the system performance. The quality of the signal enhancement was evaluated with regard to sensitivity and noise level in the acquired spectra. The microstructured Si IREs gave a signal enhancement of up to a factor of 3.8 compared to a basic Si element, with some concomitant increase in noise. The absorbance was higher for both types of Si IREs as compared to the diamond ATR crystal. The effective enhancement and the limit of quantification improved by a factor up to 3.1 in the signal-enhanced IREs compared to the basic Si element.

Mid-infrared spectroscopy using multivariate analysis for quantification has high potential in biosensing. This case study on aqueous glucose solutions yields improved prediction errors by optimising preprocessing and wavelength selection procedures.