Improving X-ray analytics: new publication in Nano Letters!

Just out in Nano Letters is our 100% lab-driven work on: Leveraging Machine Learning for Advanced Nanoscale X-ray Analysis: Unmixing Multicomponent Signals and Enhancing Chemical Quantification

In this letter, LSME introduces a new method for processing STEM-EDX spectroscopy data sets, that we term non-negative matrix factorization based pan-sharpening (PSNMF). Leveraging the Poisson nature of EDX spectral noise and binning operations, PSNMF retrieves high-quality phase spectral and spatial signatures via consecutive factorizations. At the same time, it generates reconstructed data sets with superior denoising than that provided by the “gold standard” approach of principal component analysis.