Statistical Modelling Investigation of MALDI-MSI-Based Approaches for Document Examination

Area of Science:

• Forensic Science
• Analytical Chemistry
• Spectroscopy

Background:

• Questioned document examination is crucial for forgery detection.
• Spectroscopy methods are the standard for document analysis.
• Matrix-Assisted Laser Desorption Ionisation-Mass Spectrometry Imaging (MALDI-MSI) is a powerful tool for analyzing biological traces.

Purpose of the Study:

• To explore the application of MALDI-MSI in questioned document examination.
• To differentiate between seven gel pens based on ink composition using MALDI-MSI and chemometrics.
• To assess the stability and classification robustness of ink analysis over 44 days.

Main Methods:

• Utilized Matrix-Assisted Laser Desorption Ionisation-Mass Spectrometry Imaging (MALDI-MSI) for ink chemical composition detection and imaging.
• Employed chemometric approaches, including elastic net logistic regression, for data modeling and classification.
• Validated the classification model using blind signature analysis and machine learning cross-validation.

Main Results:

• MALDI-MSI successfully detected and imaged the chemical composition of gel pen inks.
• The combined MALDI-MSI and chemometric approach achieved 100% accuracy in classifying different pens.
• The classification model demonstrated robustness in differentiating inks over a 44-day period.

Conclusions:

• The coupling of MALDI-MSI with machine learning provides a robust method for ink discrimination in questioned documents.
• This technique shows significant potential for expanding forensic versatility in document analysis.
• Further research is warranted to investigate confounding factors like paper types and environmental conditions.