Prediction of specialty coffee flavors based on near-infrared spectra using machine- and deep-learning methods.

BACKGROUND: Specialty coffee fascinates people with its bountiful flavors. Currently, flavor descriptions of specialty coffee beans are only offered by certified coffee cuppers. However, such professionals are rare, and the market demand is tremendous. The hypothesis of this study was to investigate the feasibility to train machine learning (ML) and deep learning (DL) models for predicting the flavors of specialty coffee using near-infrared spectra of ground coffee as the input. Successful model development would provide a new and objective framework to predict complex flavors in food and beverage products. RESULTS: In predicting seven categories of coffee flavors, the models developed using the ML method (i.e. support vector machine) and the deep convolutional neural network (DCNN) achieved similar performance, with the recall and accuracy being 70-73% and 75-77% respectively. Through the proposed visualization method - a focusing plot - the potential correlation among the highly weighted spectral region of the DCNN model, the predicted flavor categories, and the corresponding chemical composition are presented. CONCLUSION: This study has proven the feasibility of applying ML and DL methods on the near-infrared spectra of ground coffee to predict specialty coffee flavors. The effective models provided moderate prediction for seven flavor categories based on 266 samples. The results of classification and visualization indicate that the DCNN model developed is a promising and explainable method for coffee flavor prediction. © 2021 Society of Chemical Industry.

Determining sucrose and glucose levels in dual-purpose sorghum stalks by Fourier transform near infrared (FT-NIR) spectroscopy.

BACKGROUND: Sorghum is an advanced biomass feedstock from which grain, sugar and stover can be used for biofuel production. Determinations of specific sugar contents in sorghum stalks help to make strategic decisions during plant breeding, processing, storage and optimization of fermentation conditions. In this study, Fourier transform near infrared (FT-NIR) spectroscopy was used as a relatively fast, low-cost, high-throughput assay to predict sucrose and glucose levels in stalks of 40 dwarf grain sorghum inbreds. RESULTS: The diffuse reflection spectra were pretreated with multiplicative scatter correction (MSC) and first-derivative Savitzy-Golay (SG-1). Calibrated models were developed by partial least squares regression (PLSR) analysis. Martens' uncertainty test was used to determine the most effective spectral region. The PLSR model for stalk sucrose content was built on 380 significant wavenumbers in the 4000-6999 cm(-1) range. The model was based on four factors and had RPD = 2.40, RMSEP = 1.77 and R(2) = 0.81. Similarly, the model for stalk glucose was built using 4000-9000 cm(-1) and six factors, with RPD = 2.45, RMSEP = 0.73 and R(2) = 0.81. CONCLUSION: PLSR models were developed based on FT-NIR spectra coupled with multivariate data analysis to provide a quick and low-cost estimate of specific sugar contents in grain sorghum stalks. This sugar information helps decision making for sorghum-based biomass processing and storage strategies.

Decomposition and solubility of H2O2: implications in exhaled breath condensate.

Hydrogen peroxide (H2O2) is one of the metabolic end products present in exhaled breath. High levels of H2O2 found in breath condensate are an indicator of airway inflammation and could be used for monitoring the condition of patients with chronic obstructive pulmonary disease. However, sampling conditions such as breath temperature, condensing temperature, flow rate and collection time can affect the intrinsic properties of H2O2-its solubility, volatility, and decomposition rate. Sudden decreases to H2O2 concentration may be due to the sampling conditions instead of the patient's health status. The decomposition rate and Henry's law constant for saturated H2O2 vapor (RH > 95%) within 22-42 °C, which correlates to room temperature and range of human breath temperatures, are needed for better understanding and standardization of breath collection. In this study, we determined the effects of initial H2O2 concentration, temperature, and sampling time on the decomposition rate by comparing electrochemical measurements of H2O2 in simulated breath samples. The experimental results showed the decomposition rate of H2O2 increased as the breath temperature and sampling time increased and the solubility of H2O2 increased with increasing flow rate and condensing temperature during sampling. Prediction models for H2O2 sensing in exhaled breath sample were developed that could be used in the standardization of exhaled breath condensate collection. These experimental findings need to be further verified with human/animal breath samples.