GMS 3 Analysis Tools: Multivariate Statistical Analysis
Electron energy-loss spectrum imaging is well established as a powerful tool for materials analysis. The wealth of information available beyond simple composition, coupled with high signal collection efficiency in next-generation imaging filters makes this technique particularly advantageous for advanced analysis and mapping.
Multivariate analysis methods are becoming increasingly popular for spectrum image analysis due to the capability these methods offer for blind analysis. Principle component analysis (PCA), one of the first and most established approaches to multivariate data analysis, organizes multivariate data into a projection in which the total variance of the projected data is maximized, which can greatly improve the interpretability of the data.
Gatan Microscopy Suite (GMS) 3.4 includes a new multivariate statistical analysis tool that includes several key features such as single-click denoise, PCA decomposition, and varimax factor rotation. The new software release also includes Python integration, giving the exciting opportunity to perform multivariate analysis using external Python libraries and packages in addition to the easy-to-use multivariate statistical analysis (MSA) tool already included.
This webinar provides a detailed overview of the new MSA tool in addition to several MSA worked examples using external Python packages to highlight the new data processing capabilities.