Unsupervised Deep Learning for Limited-Angle STEM-EDX Tomography -- Application to 3D Chemical Analysis of Phase-Change Memory Devices
arXiv:2606.10547v1 Announce Type: cross Abstract: Energy Dispersive X-ray (EDX) tomography in Scanning Transmission Electron Microscopy (STEM) enables 3D compositional and elemental mapping at the nanoscale, but its use is limited by restricted tilt ranges and low-dose conditions required to avoid beam damage. Limited-angle acquisition introduces missing-wedge artefacts such as elongation and anisotropic resolution, while noisy low-dose data further degrade reconstruction quality and quantitative reliability. Here, we introduce an unsupervised deep learning framework based on Deep Image Prior
The convergence of advanced imaging techniques like STEM-EDX with increasingly sophisticated unsupervised deep learning methods is enhancing capabilities for material analysis.
This development improves resolution and reduces artifacts in nanoscale imaging, critical for advancing materials science, particularly in semiconductor and memory technologies.
The ability to perform more accurate 3D chemical analysis with limited data acquisition fundamentally changes how researchers and engineers characterize novel materials and devices.
- · Materials Science Researchers
- · Semiconductor Industry
- · Deep Learning Algorithm Developers
- · Advanced Microscopy Manufacturers
- · Traditional Materials Characterization Methods (without AI integration)
Improved understanding of material properties at the atomic scale facilitates the development of next-generation devices.
Faster design cycles for new materials and devices due to more efficient and accurate characterization processes.
Acceleration of innovation in areas requiring precise material engineering, such as quantum computing and advanced energy storage.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG