Description

The Electron Microscopy of Materials (EMM) unit at Linköping University invites applications for a PhD position at the intersection of AI and advanced electron microscopy. The project focuses on developing novel self-supervised and physics-informed deep learning methods to restore and denoise Transmission Electron Microscopy (TEM) and HAADF-STEM images.

Microscopy data are often degraded by noise and scan distortions, and clean ground truth data are rarely available. This project aims to go beyond standard convolutional neural networks by exploring transformers, implicit neural representations (INRs), and hybrid architectures that integrate physical priors such as periodicity, symmetry, and long-range correlations. The goal is to achieve more robust, interpretable, and scientifically meaningful reconstructions of atomic-resolution microscopy data.

This position is part of a cross-disciplinary effort supported by the Wallenberg Initiative Materials Science for Sustainability (WISE) and the Wallenberg AI, Autonomous Systems and Software Program (WASP), providing a unique opportunity to work within a strong Swedish research network spanning materials science, AI, and computational imaging.

Submission is possible until: 17 October 2025

Requirements

• Master’s degree in Computer Science, Materials Science, Physics, Mathematics, or related fields
• Strong interest in AI for scientific imaging, self-supervised learning, and image restoration
• Experience in deep learning (CNNs, transformers, or INRs) and scientific programming
• Good analytical and communication skills, with the ability to work independently and collaboratively
• Experience with microscopy data or inverse problems is a merit

More information

Go to the job offer page

Contact

Anna Elsukova

Principal Research Engineer, Electron Microscopy of Materials, Linköping University

[email protected]

Per Persson

Professor, Head of Unit, Electron Microscopy of Materials, Linköping University

[email protected]