HREM Research has released its Multivariate Statistical Analysis (MSA) plug-in for DigitalMicrograph. Launched at the M&M2008 show in Albuquerque, USA, and at EMC2008 in Aachen, Germany, MSA is an advanced tool for the analysis of electron microscope spectrum images (SIs), and works with Gatan's DigitalMicrograph as a plug-in. MSA finds statistically significant features buried under a heavy statistical noise from 2D and 3D spectrum images gathered by spectrometric techniques such as XEDS, EELS, EFTEM and cathodoluminescence.
MSA was originally developed by Masashi Watanabe mainly at Lehigh University. It features principal component analysis (PCA) and automatically extracts statistically significant spectral features. It also includes several utilities to handle SIs efficiently and to import spectra obtained by other acquisition systems than Gatan DigitalMicrograph.
"A spectrum image is extremely redundant, since each component spectrum spans the same energy range, and contains the same features specific to the element in the sample," said Dr Kazuo Ishizuka, President of HREM Research Inc. "Thus, we can extract statistically significant information buried under extremely heavy noise without degrading spatial information, as we can see from an attached figure. Every SI data should be processed in this way to extract hidden information."
Picture caption: Raw and MSA-reconstructed Cu maps in irradiated low-alloy steel.
About HREM Research Inc.
Founded in 2001, HREM Research Inc. specializes in developing products and services that enhance High-Resolution Electron Microscopy (HREM). Dr. Kazuo Ishizuka, the founder of a company has established the whole technique for HREM image simulation. Thus, a company's flagship product, Mac/WinHREM, is a world leading HREM image simulation software. Currently, HREM Research Inc. is actively working on making useful techniques to be available for the HREM community. For more information, visit http://www.hremreserch.com or contact support@hremreserch.com.
