Analysis of Allergenic Pollen by FTIR Microspectroscopy

Area of Science:

• Aerobiology
• Spectroscopy
• Chemometrics

Background:

• Fourier transform infrared (FTIR) spectroscopy is valuable for pollen and spore identification.
• Mie-type scattering complicates spectral interpretation and multivariate analysis of single pollen grains.
• Existing methods for automated pollen analysis have limitations in accuracy and efficiency.

Purpose of the Study:

• To develop a novel sampling setup for infrared microspectroscopy that minimizes Mie-type scattering.
• To establish a robust classification model for pollen identification using sparse partial least-squares regression (PLSR).
• To compare the performance of the new method with traditional optical microscopy.

Main Methods:

• Pollen samples were embedded in paraffin between polyethylene foils, avoiding pretreatment.
• Single-grain infrared spectra were acquired and analyzed using extended multiplicative signal correction and sparse PLSR.
• Classification models were built and validated using training (260 spectra) and independent test sets (130 spectra).

Main Results:

• The novel sampling setup effectively prevented strong Mie-type scattering.
• Sparse PLSR generated robust classification models with biochemical interpretability.
• The method achieved high accuracy rates of 95% on the independent test set and 98% with full cross-validation.
• The infrared microspectroscopy method demonstrated superior taxonomic resolution, speed, economy, and bias-free measurement compared to optical microscopy.

Conclusions:

• The developed infrared microspectroscopy method with a novel sampling setup provides accurate and efficient pollen and spore identification.
• The technique is compatible with standard air-samplers, facilitating its use in routine aerobiology studies.
• This approach offers significant advantages over optical microscopy, including enhanced taxonomic resolution and cost-effectiveness.