InSituAnalyze: A Python Framework for Multicomponent Synchronous Analysis of Spectral Imaging

Area of Science:

• Spectral imaging
• Multicomponent analysis
• Data visualization

Background:

• Spectral imaging offers high precision and sensitivity for complex material analysis.
• High-dimensional data from spectral imaging requires advanced feature extraction and information mining.
• Analyzing spatial distribution of materials necessitates efficient data processing.

Purpose of the Study:

• To develop a user-friendly Python framework for multicomponent synchronous analysis of spectral imaging data.
• To enable semiquantitative analysis of target components on a pixel scale for complex samples.
• To facilitate efficient handling and utilization of high-dimensional spectral imaging data.

Main Methods:

• Development of a Python framework utilizing a characteristic band method.
• Integration of a fast Non-Negative Least Squares (fast-NNLS) algorithm for spectral unmixing.
• Implementation of user-selectable pretreatment methods for images and spectra.

Main Results:

• The framework enables intuitive and time-saving analysis of spectral imaging data.
• It allows for the extraction of spatial information from multispace data, including tissues and structures.
• Semiquantitative information on target components is obtained on a pixel-by-pixel basis.

Conclusions:

• The developed Python framework simplifies complex spectral imaging data analysis.
• It provides a powerful tool for researchers needing semiquantitative insights into material composition.
• The framework's architecture supports extensibility for new algorithms and data formats.