Electron Microscopy
Electron microscopy (EM) generates high-resolution images of materials at the nano- and atomic scale, crucial for characterizing diverse materials and biological structures. Current research emphasizes developing advanced computational methods, including deep learning architectures like convolutional neural networks (CNNs), generative adversarial networks (GANs), and hypergraph neural networks (HgNNs), to automate image analysis, improve segmentation accuracy (especially with limited labeled data), and enhance image resolution through super-resolution techniques. These advancements are significantly impacting materials science, biology, and medicine by enabling faster, more efficient, and higher-throughput analysis of complex datasets, leading to improved material design and a deeper understanding of biological systems.
Papers
Using convolutional neural networks for stereological characterization of 3D hetero-aggregates based on synthetic STEM data
Lukas Fuchs, Tom Kirstein, Christoph Mahr, Orkun Furat, Valentin Baric, Andreas Rosenauer, Lutz Maedler, Volker Schmidt
Deep Learning Enables Large Depth-of-Field Images for Sub-Diffraction-Limit Scanning Superlens Microscopy
Hui Sun, Hao Luo, Feifei Wang, Qingjiu Chen, Meng Chen, Xiaoduo Wang, Haibo Yu, Guanglie Zhang, Lianqing Liu, Jianping Wang, Dapeng Wu, Wen Jung Li
Deep Learning for Automated Experimentation in Scanning Transmission Electron Microscopy
Sergei V. Kalinin, Debangshu Mukherjee, Kevin M. Roccapriore, Ben Blaiszik, Ayana Ghosh, Maxim A. Ziatdinov, A. Al-Najjar, Christina Doty, Sarah Akers, Nageswara S. Rao, Joshua C. Agar, Steven R. Spurgeon
FakET: Simulating Cryo-Electron Tomograms with Neural Style Transfer
Pavol Harar, Lukas Herrmann, Philipp Grohs, David Haselbach