A recent study demonstrated how near-infrared (NIR) spectroscopy can predict adulteration in apricot kernels with high accuracy.
According to a recent study published in the journal Food Control, near-infrared (NIR) spectroscopy is an effective technique in identifying adulteration of ground almonds (1).
Almonds are one of the most popular nuts globally and contain high nutritional content. As a result, they are a popular food for many consumers, and are often used as a reliable protein source. However, like many foods, almonds are susceptible to food adulteration, which is a growing problem globally. Food production companies and other actors in the food industry have great incentive for altering the ingredients in the food we consume. In many cases, food adulteration results in the use of new chemicals in food that help make the food more addictive and unhealthier.
Wooden spoon with almond meal | Image Credit: © Picture Partners - stock.adobe.com
As a result, detecting adulteration in ground almonds, particularly with apricot kernels, is important. Traditional methods like chromatography, while effective, are costly and destructive (1). Currently, many spectroscopic techniques are being used for this purpose instead. These techniques include hyperspectral imaging (HSI), Raman spectroscopy, nuclear magnetic resonance (NMR) spectroscopy, and fluorescence (FL) spectroscopy (2).
In this study, a multidisciplinary research team comprised of researchers from Agri Ibrahim Cecen University, University of Cordoba, Sivas Science and Technology University, and Canakkale Onsekiz Mart University investigated the potential of using near-infrared (NIR) spectroscopy for identifying almond adulteration. The theory was that NIR spectroscopy, because of its ability to conduct rapid and nondestructive analysis of samples, would be successful in detecting the level of food adulteration in ground almonds (1).
To test their theory, the research team collected 120 almond samples from around Turkey. Then, to gather the NIR spectra, the research team used benchtop and portable spectrometers on each sample (1). Next, the research team had to classify and authenticate ground almonds using machine learning algorithms.
Machine learning (ML) is an important tool in detecting food adulteration. ML algorithms can identify complex patterns in large data sets that are difficult or impossible for humans to detect, offering an automated solution (2). ML can handle and process these large data sets efficiently, providing insights that traditional methods might miss. These ML models can also quickly analyze data and provide accurate results, which is crucial for timely detection of food adulteration (2).
The researchers in this study used soft independent modeling of class analogy (SIMCA) and conditional entropy (CE) with ML algorithms to classify and authenticate ground almonds (1). To predict the apricot kernel levels in adulterated samples, partial least squares regression (PLSR) and CE were used (1).
Between the SIMCA and CE algorithms with the spectral data, the researchers demonstrated the following: These algorithms were able to achieve 100% accuracy in distinguishing between pure and adulterated almond samples (1). The researchers also determined that both portable and benchtop spectrometers performed similarly. They observed high correlation values (rval > 0.96) and a standard error prediction (SEP) of 3.98% for portable units and SEP>4.49 for benchtop units (1).
As a result, the findings showed that SIMCA, PLSR, and CE models can be used in combination with NIR spectroscopy to detect apricot kernel adulteration in ground almonds. The study, therefore, proved the initial theory correct in that NIR spectroscopy, with machine learning algorithms, can be used to detect adulteration in apricot kernels in ground almonds.
(1) Menevseoglu, A.; Entrenas, J. A.; Gunes, N.; et al. Machine Learning-assisted Near-Infrared Spectroscopy for Rapid Discrimination of Apricot Kernels in Ground Almond. Food Cont. 2024, 159, 110272. DOI: 10.1016/j.foodcont.2023.110272
(2) Goyal, R.; Singha, P.; Singh, S. K. Spectroscopic Food Adulteration Detection Using Machine Learning: Current Challenges and Future Prospects. Trends Food Sci. Technol. 2024, 146, 104377. DOI: 10.1016/j.tifs.2024.104377
Artificial Intelligence Accelerates Molecular Vibration Analysis, Study Finds
July 1st 2025A new review led by researchers from MIT and Oak Ridge National Laboratory outlines how artificial intelligence (AI) is transforming the study of molecular vibrations and phonons, making spectroscopic analysis faster, more accurate, and more accessible.
Toward a Generalizable Model of Diffuse Reflectance in Particulate Systems
June 30th 2025This tutorial examines the modeling of diffuse reflectance (DR) in complex particulate samples, such as powders and granular solids. Traditional theoretical frameworks like empirical absorbance, Kubelka-Munk, radiative transfer theory (RTT), and the Hapke model are presented in standard and matrix notation where applicable. Their advantages and limitations are highlighted, particularly for heterogeneous particle size distributions and real-world variations in the optical properties of particulate samples. Hybrid and emerging computational strategies, including Monte Carlo methods, full-wave numerical solvers, and machine learning (ML) models, are evaluated for their potential to produce more generalizable prediction models.
Polystyrene and UVC Sterilization Tested with Spectroscopy and Luminescence Tools
June 25th 2025A team of researchers from Spanish institutions has found that polystyrene used in healthcare packaging shows strong resistance to UVC sterilization, with minimal chemical degradation detected using FT-IR and Raman spectroscopy.