A FAIR Data Ecosystem in Transmission Electron Microscopy

This thesis, titled "A FAIR Data Ecosystem in Transmission Electron Microscopy", documents the development of a web-based data management platform at the Laboratory of Electron Microscopy (LAME) in Area Science Park, Trieste, designed to enforce FAIR (Findable, Accessible, Interoperable, Reusable) principles from data acquisition through to archival. Centered around the JEOL JEM-F200 TEM/STEM system, the platform enables the structured capture, transformation, and storage of complex electron microscopy data—particularly TIFF-based TEM and SAED images—into NeXus NXem-compliant HDF5 containers. Built with Django and deployed via Kubernetes, the system features automated metadata extraction from image headers, manual metadata injection via user-friendly web forms, and seamless integration with MinIO object storage and Authentik identity management for secure, scalable access. Through rigorous NeXus validation and policy-aligned naming conventions under the NFFA-DI infrastructure, this work not only enhances internal data quality and reproducibility at LAME but also contributes a replicable model for FAIR-aligned data stewardship in microscopy-focused research environments.