Rapid qualitative detection of titanium dioxide adulteration in persimmon icing using portable Raman spectrometer and...

Area of Science:

• Food Science and Technology
• Analytical Chemistry
• Spectroscopy

Background:

• Persimmon icing is a natural byproduct of persimmon cake processing.
• Titanium dioxide is illicitly used to mimic high-quality persimmon icing, posing health risks.
• Accurate detection methods are crucial for food safety and quality control.

Purpose of the Study:

• To develop and validate a rapid, reliable method for detecting titanium dioxide adulteration in persimmon icing.
• To evaluate the efficacy of Raman spectroscopy combined with machine learning algorithms for this purpose.

Main Methods:

• Raman spectroscopy was employed for label-free, in-situ analysis of persimmon icing.
• The adaptive iterative reweighting partial least squares (air-PLS) algorithm was used for fluorescent background correction.
• Principal Components Analysis (PCA) was used for spectral data analysis.
• Various machine learning models, including extreme learning machine (ELM), support vector machine (SVM), back propagation artificial neural network (BP-ANN), and random forest (RF), were compared.
• A one-dimensional stack auto encoder convolutional neural network (1D-SAE-CNN) was developed and optimized.

Main Results:

• Principal Components Analysis (PCA) effectively explained 99.9% of the spectral data variance using the first two principal components (PC-1 and PC-2).
• The 1D-SAE-CNN model achieved the highest performance metrics: detection accuracy of 0.9825, precision of 0.9824, recall of 0.9825, and F1-score of 0.9824.
• The developed method demonstrated high efficacy in distinguishing between authentic and adulterated persimmon icing.

Conclusions:

• Raman spectroscopy integrated with machine learning, particularly the 1D-SAE-CNN model, offers a highly accurate and promising approach for detecting titanium dioxide adulteration in persimmon icing.
• This technique provides a rapid, label-free, and field-deployable solution for ensuring the safety and quality of persimmon products.
• The findings support the adoption of this advanced analytical method in food quality assurance protocols.