Conventional versus AI-based spectral data processing and classification approaches to enhance LIBS's analytical...

Area of Science:

• Forensic Science
• Analytical Chemistry
• Spectroscopy

Background:

• Laser-Induced Breakdown Spectroscopy (LIBS) is a valuable technique for material analysis.
• Conventional methods for LIBS data processing, such as Principal Component Analysis (PCA) and Partial Least Squares Discriminant Analysis (PLS-DA), often require extensive user preprocessing.
• There is a need for more efficient and accurate methods for LIBS data interpretation, particularly in forensic applications like sample discrimination.

Purpose of the Study:

• To develop and evaluate a novel Artificial Intelligence (AI)-driven approach for processing and interpreting Laser-Induced Breakdown Spectroscopy (LIBS) data.
• To compare the performance of the AI-developed method against conventional techniques (PCA, PLS-DA) for forensic toner sample discrimination.
• To demonstrate the potential of AI in enhancing the accuracy and efficiency of spectroscopic analysis for forensic applications.

Main Methods:

• A novel AI-developed method was proposed, integrating normalization, interpolation, and peak detection for automated LIBS spectral analysis.
• The AI method was compared with conventional Principal Component Analysis (PCA) and Partial Least Squares Discriminant Analysis (PLS-DA).
• Performance evaluation involved quantitative statistical analysis, including accuracy difference percentage, variance analysis, paired t-test, and cross-validation.

Main Results:

• The AI-developed method demonstrated superior performance in discriminating between toner samples from various printer and photocopier brands and models.
• The AI approach significantly improved accuracy compared to conventional LIBS data analysis methods.
• The proposed method requires no user preprocessing, simplifying the LIBS spectral analysis workflow.

Conclusions:

• Artificial Intelligence significantly enhances Laser-Induced Breakdown Spectroscopy for forensic applications, particularly in sample discrimination.
• The novel AI-developed method offers increased efficiency and accuracy in analyzing LIBS data, eliminating the need for user preprocessing.
• This study highlights the transformative potential of AI in advancing spectroscopic techniques for forensic science and other analytical fields.