Materials Science

Energy-Dispersive X-ray (EDX) Data Analysis

Energy-dispersive x-ray spectroscopy produces high-dimensional data that characterizes the elemental composition of materials at the microscale. We develop multiscale Bayesian algorithms for the analysis of EDX data, enabling robust estimation of elemental maps and composition profiles. Our approaches account for the Poisson noise statistics inherent in photon-counting data and exploit spatial correlations through hierarchical models for improved estimation accuracy.