A recent study developed an accurate, non-destructive geo-traceability method using NIR spectroscopy and machine learning to authenticate the geographic origins of Gastrodia elata Bl.
In a recent study published in Food Chemistry, researchers from Yunnan Academy of Agricultural Sciences, Yunnan Agricultural University, and Zhaotong University developed a non-destructive and digital geo-traceability model using near-infrared (NIR) spectroscopy combined with advanced chemometric approaches (1). This model was used to improve the evaluation of the authenticity and quality of Gastrodia elata Bl (GE).
Wild gastrodia elata | Image Credit: © feng67 - stock.adobe.com.
GE is an herbal medicine that contains many therapeutic properties (2). It is often used to treat various neurological disorders (2). For example, GE has been used to treat headaches and migraines (3). Also known as Tian Ma, GE has a unique chemical composition and active ingredient content, but these two variables can be influenced by environmental factors (1). These environmental factors include the quality of the soil and the climate (1). Therefore, geographical traceability methods are needed to ensure the authenticity and quality of GE to make sure the product is safe for consumers (1).
In this study, the research team focused on addressing this challenge by leveraging the power of NIR spectroscopy and machine learning (ML). The approach developed in the study was designed to make it easy for regulators and stakeholders in the herbal medicine industry to quickly authenticate GE sources with better accuracy (1).
The study utilized a multi-faceted strategy to establish the geotraceability of GE. For sample collection, the researchers collected GE from four select regions in China: Zhaotong Xiaocaoba in Yunnan; Dafang County in Guizhou; Bomi County in Xizang; and Wufeng County in Hubei (1). These regions were selected to capture geographic variability in the herb’s chemical composition (1).
Then, NIR spectroscopy was used to analyze the chemical profiles of the samples. To create the ML models, several advanced chemometric methods were used. These methods include partial least squares–discriminant analysis (PLS-DA), genetic algorithm–support vector machine (GS-SVM), backpropagation neural network (BPNN), gradient boosting machine (GBM), and residual network (ResNet) (1).
A ResNet model was built using two-dimensional (2D) and three-dimensional (3D) correlation spectroscopy (COS) images generated from the NIR spectroscopy data (1). The performance of the models was validated using a separate data set from Zhaotong City, Yunnan. The 3D COS-ResNet model achieved great results, with 100% accuracy on the test set and 95.45% external validation accuracy, outperforming traditional ML models (1).
The geotraceability method is not only accurate, but it was also non-destructive and eco-friendly. As a result, the researchers show that their method was good at ensuring minimal waste and no impact on the analyzed samples (1). By digitizing the authentication process, the method also reduces the reliance on time-intensive and costly traditional testing techniques (1).
The integration of ML and spectroscopy offers a scalable and efficient solution for addressing fraud and quality control issues in the herbal medicine industry. Despite the promising results, the researchers emphasized the need for further studies to enhance the model’s robustness and applicability (1). Expanding the data set to include more geographic regions and exploring additional chemometric techniques are among the proposed next steps outlined in the study (1). As the herbal medicine industry continues to grow, innovations like this will be instrumental in ensuring quality and authenticity while safeguarding consumer confidence.
Exoplanet Discovery Using Spectroscopy
March 26th 2025Recent advancements in exoplanet detection, including high-resolution spectroscopy, adaptive optics, and artificial intelligence (AI)-driven data analysis, are significantly improving our ability to identify and study distant planets. These developments mark a turning point in the search for habitable worlds beyond our solar system.
New Telescope Technique Expands Exoplanet Atmosphere Spectroscopic Studies
March 24th 2025Astronomers have made a significant leap in the study of exoplanet atmospheres with a new ground-based spectroscopic technique that rivals space-based observations in precision. Using the Exoplanet Transmission Spectroscopy Imager (ETSI) at McDonald Observatory in Texas, researchers have analyzed 21 exoplanet atmospheres, demonstrating that ground-based telescopes can now provide cost-effective reconnaissance for future high-precision studies with facilities like the James Webb Space Telescope (JWST) (1-3).