Evaluation of rice varieties using LIBS and FTIR techniques associated with PCA and machine learning algorithms

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Chemometrics

Background:

• Laser-induced breakdown spectroscopy (LIBS) and Fourier transform infrared spectroscopy (FTIR) are versatile spectroscopic techniques.
• These methods offer elemental and molecular characterization, respectively.
• Accurate rice differentiation is crucial for quality control and nutritional assessment.

Purpose of the Study:

• To evaluate LIBS and FTIR combined with Principal Component Analysis (PCA) and Machine Learning (ML) for rice composition analysis.
• To identify key chemical features responsible for differentiating rice varieties.
• To assess the accuracy and effectiveness of spectroscopic methods for rice classification.

Main Methods:

• Elemental analysis using LIBS.
• Molecular identification using FTIR.
• Data analysis with PCA and supervised ML algorithms.
• Classification of white, brown, black, and red rice samples.

Main Results:

• PCA identified protein, fatty acids, and magnesium as key differentiators between rice types.
• ML analysis achieved 100% accuracy, sensitivity, and specificity in sample classification.
• Spectroscopic data revealed subtle chemical variations crucial for differentiation.

Conclusions:

• LIBS and FTIR coupled with multivariate analysis are effective for rice composition analysis and differentiation.
• These techniques offer a promising alternative to traditional analytical methods.
• The combined approach provides highly accurate classification of rice varieties.