Rapid Identification of Rainbow Trout Adulteration in Atlantic Salmon by Raman Spectroscopy Combined with Machine Learning

Abstract

This study intends to evaluate the utilization potential of the combined Raman spectroscopy and machine learning approach to quickly identify the rainbow trout adulteration in Atlantic salmon. The adulterated samples contained various concentrations (0–100% w/w at 10% intervals) of rainbow trout mixed into Atlantic salmon. Spectral preprocessing methods, such as first derivative, second derivative, multiple scattering correction (MSC), and standard normal variate, were employed. Unsupervised algorithms, such as recursive feature elimination, genetic algorithm (GA), and simulated annealing, and supervised K-means clustering (KM) algorithm were used for selecting important spectral bands to reduce the spectral complexity and improve the model stability. Finally, the performances of various machine learning models, including linear regression, nonlinear regression, regression tree, and rule-based models, were verified and compared. The results denoted that the developed GA–KM–Cubist machine learning model achieved satisfactory results based on MSC preprocessing. The determination coefficient (R2) and root mean square error of prediction sets (RMSEP) in the test sets were 0.87 and 10.93, respectively. These results indicate that Raman spectroscopy can be used as an effective Atlantic salmon adulteration identification method; further, the developed model can be used for quantitatively analyzing the rainbow trout adulteration in Atlantic salmon.

A Machine Learning Method for the Quantitative Detection of Adulterated Meat Using a MOS-Based E-Nose

“…However, they require complex extractions and have long analysis times which significantly limit their widespread use [ 14 ]. For spectroscopy-based technologies, near-infrared spectroscopy (NIRS) [ 15 ], Raman spectroscopy (RS) [ 16 ], and hyperspectral imaging (HSI) [ 17 ] have been shown to be useful for detecting meat adulteration. However, the complex spectroscopy data requires a high degree of technical expertise to analyze.…”

Classification of pathogens by Raman spectroscopy combined with generative adversarial networks

“…At present, most Raman spectroscopic identification methods for blood products are linear models (Chen et al, 2019b); however, due to stray light in the instrument, ambient temperature effects and the nonlinear response of the detector, the spectral response to a substance is not completely linear, which can lead to poor classification results. Although the SVM method has been used as a nonlinear identification method (Lussier et al, 2019), it also has some problems, such as poor interpretability.…”

Fraud in Animal Origin Food Products: Advances in Emerging Spectroscopic Detection Methods over the Past Five Years

“…Authentication in meats is required for foods that are labeled as individual meats [71]. Horsemeat in minced beef [72], beef and mutton in pork [73], and rainbow trout in Atlantic salmon [74] each require sufficient data specific to substances to be labeled to assure the meat contaminant is properly characterized in order to identify markers characteristic of each additional component. Spectral data on the primary meat preferentially needs to be oversampled relative to that of a contaminant, or of minor or occasional components that could be misinterpreted as unrelated to the original meat.…”

“…Using PCA, the authors succeeded in discriminating 11 formulations with different percentages of these two species, and the percentage of the fraud in the mixture was successfully predicted using PLSR. The same authenticity issue (i.e., species identification) was later studied in a similar investigation, but with a different vibrational spectroscopic technique, namely Raman spectroscopy [74].…”