Critical Review of Processing and Classification Techniques for Images and Spectra in Microplastic Research

Area of Science:

• Environmental Science
• Analytical Chemistry
• Materials Science

Background:

• Microplastic research demands high-throughput, automated analysis techniques.
• Current methods for microplastic identification often involve optical and spectral analysis, with varying levels of automation.
• Standardized protocols for data processing and interpretation are needed to ensure reliable results.

Purpose of the Study:

• To review and critique current image and spectral analysis techniques for microplastics.
• To identify opportunities for advancing microplastic data analysis and interpretation.
• To propose strategies for developing high-throughput screening methods.

Main Methods:

• Comprehensive review of image analysis techniques (optical microscopy, SEM, fluorescence microscopy).
• Review of spectral analysis techniques (FT-IR, Raman, Py-GC-MS, EDX).
• Critique of data processing and classification approaches for both image and spectral data.

Main Results:

• Microplastic research is increasingly integrating multiple analytical methods and chemical classification.
• Opportunities exist in quantifying particle characteristics, advancing spectral processing, and creating robust spectral libraries.
• Machine learning models show promise for automated microplastic classification.

Conclusions:

• Advancements in microplastic analysis require improved image and spectral processing, robust reference libraries, and rigorous validation.
• Developing and sharing analytical tools, raw data, and code are essential for progress.
• Machine learning offers a promising avenue for achieving high-throughput microplastic screening.