Researchers have developed a rapid and accurate method that combines front-excited emission matrix fluorescence spectroscopy with interpretable deep learning to identify the storage year of Ningxia wolfberry and put it into the environment to combat market fraud. offered a friendly solution.
To combat the deceptive practice of selling aged Ningxia wolfberry as fresh food, researchers at Hunan University in Changsha, China, have developed a new method to quickly identify the shelf life of this precious fruit. developed. their research Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopypresents a combination of front-excited emission matrix (FF-EEM) fluorescence spectroscopy and interpretable deep learning as a powerful tool for this purpose (1).
FF-EEM fluorescence spectroscopy A technique in which a sample is illuminated with different wavelengths of light and the corresponding emitted fluorescence is measured. It provides a comprehensive spectral fingerprint of the sample, capturing the unique fluorescence properties of the sample. Valuable chemical information about a sample can be extracted by analyzing excitation and emission spectra using deep learning. Deep learning is a branch of machine learning that uses artificial neural networks to learn from complex data and make predictions.
Known for its antioxidant and health benefits, Ningxia wolfberry fetches a high price in the market. However, unscrupulous traders may attempt to disguise old goji berries as freshly harvested, deceiving consumers and potentially causing financial loss. The proposed method aims to address this challenge by providing a lossless, fast and accurate means of determining the preservation year of Ningxia wolfberry.
The researchers used the alternating trilinear decomposition (ATLD) algorithm to extract chemically significant information from the three-way data array obtained from the Ningxia wolfberry sample. This pretreatment step allowed us to isolate key features associated with fruit shelf life. In addition, the research team introduced a convolutional neural network (CNN) model, called EEMnet, specifically designed for custody age determination. EEMnet exploits subtle spectral differences between wolfberry samples to achieve robust classification.
Remarkably, the EEMnet model showed an accurate classification rate of over 98% for the training set, test set and prediction set, demonstrating its effectiveness in distinguishing wolfberry samples of different storage years. To increase transparency and interpretability, researchers conducted a series of analyzes to elucidate the inner workings of deep learning models and provide insight into the decision-making process.
The results of this study demonstrate that the combined approach of FF-EEM fluorescence spectroscopy and EEMnet has great potential for rapid and accurate determination of the storage year of Ningxia wolfberry. This method provides a valuable contribution to determining the shelf life of herbal medicine raw materials by providing a green and reliable solution.
With the growing demand for high-quality agricultural products, the development of innovative technology combining FF-EEM fluorescence spectroscopy with interpretable deep learning paves the way for improved food authentication and increased consumer trust. Secure and protect market integrity. The results of this study could potentially be applied to other valuable food products, enabling reliable traceability and quality control across various industries.
reference
(1) Yang, X.-Q. Wu, H.-L. Wang, B. Wang, T. Chen, Y. Chen, A.-Q. Huang, K. Zhang, Y.-Y. Yang, J. Yu, R.-Q. Combining front-excited emission matrix fluorescence spectroscopy with interpretable deep learning to rapidly identify storage years in Ningxia wolfberry. Spectrochymica Acta Part A: Mol. Biomol. Spectrometer. 2023, 295, 122617. DOI: 10.1016/j.saa.2023.122617
