Wavelength selection framework for classifying food and pharmaceutical samples into multiple classes

Abstract

Near infrared (NIR) spectroscopy is an efficient, low-cost analytical technique widely applied to identify the origin of food and pharmaceutical products. NIR spectra-based classification strategies typically use thousands of equally spaced wavelengths as input information, some of which may not carry relevant information for product classification. When that is the case, the performance of predictive and exploratory multivariate techniques may be undermined by such noisy information. In this paper, we propose an iterative framework for selecting subsets of NIR wavelengths aimed at classifying samples into categories. For that matter, we integrate Principal Components Analysis (PCA) and three classification techniques: k-Nearest Neighbor (KNN), Probabilistic Neural Network (PNN) and Linear Discriminant Analysis (LDA). PCA is first applied to NIR data, and a wavelength importance index is derived based on the PCA loadings. Samples are then categorized using the wavelength with the highest index and the classification accuracy is calculated; next, the wavelength with the second highest index is inserted into the dataset and a new classification is performed. This forward-based iterative procedure is carried out until all original wavelengths are inserted into the dataset used for classification. The subset of wavelengths leading to the maximum accuracy is chosen as the recommended subset. Our propositions performed remarkably well when applied to four datasets related to food and pharmaceutical products.

A genetic algorithm‐based framework for wavelength selection on sample categorization

“…In forensic and pharmaceutical scenarios, the accurate insertion of medicine and drug samples into proper categories may unveil common and peculiar features of seized samples and guide investigative forces towards tracking illegal operations. [17][18][19][20][21] Despite the undeniable usefulness of wavelength selection, some studies have pointed out a list of precautions to be considered when tackling the issue. [16] Other interesting applications of approaches for wavelength selection can be found in the literature.…”

Similar offspring voting genetic algorithm for spectral variable selection

“…Calibration with the full range of variables is time consuming, and the irrelevant information within the spectrum would affect the accuracy and robustness of the prediction . Therefore, variable selection or uninformative variable elimination has attracted attention for the development of spectroscopy‐based calibration models …”