Webinar 24 Sep 26: AI-powered microscopy workflows for automated chemical and mineralogical analysis

The EXCITE Academy Webinar Series aims to inspire you. Experts and EXCITE users share expertise, experiences and ideas so we can advance the field together. Webinar 24 September 2026 by experts Thomas Ginnis, Postdoctoral Research Associate, and Richard Harrison, Professor of Earth and Planetary Materials, from the University of Cambridge in Cambridge, UK. Explore together with Thomas and Richard how artificial intelligence is transforming microscopy workflows for chemical and mineralogical analysis. Their work combines advanced electron microscopy with AI-powered data analysis to automate complex analytical tasks and extract detailed chemical and mineralogical information from large imaging datasets. In this webinar, they introduce the SIGMA2 workflow developed within the EXCITE Network, demonstrate how AI can be integrated into microscopy workflows, and show how these approaches can accelerate and improve the automated analysis of minerals and materials. In more detail Richard, Thomas and colleagues have developed SIGMA2, an AI-powered workflow designed to automate the analysis of microscopy data for chemical and mineralogical characterisation. By combining advanced microscopy techniques with machine learning and the HyperSpy software ecosystem, the workflow can help researchers process and interpret complex datasets more efficiently, while reducing the need for time-consuming manual analysis. The webinar showcases how SIGMA2 can be used to identify and classify phases, extract chemical information, and analyse mineralogical features within microscopy datasets. The speakers also discuss how AI-powered workflows can make advanced microscopy more scalable and reproducible, opening up new possibilities for high-throughput analysis in Earth and environmental sciences. The webinar also includes details on how EXCITE users can utilise open source tools to create customised workflows and pipelines for analysing complex microscopy datasets.